Similar to other viruses in the Coronaviridae family, SARS-CoV-2 has a single-stranded, positive-sense RNA (+ssRNA) genome of approximately 29 kb [15,16]. The viral RNA is composed of more than six open reading frames (ORFs), the first one of which (ORF1) serves as a template for producing two polyproteins essential for viral replication and transcription: pp1a and pp1ab [17,18]. These two polyproteins undergo extensive processing by the viral main protease (Mpro) and another protease known as papain-like protease (PLP), producing 16 nonstructural proteins (NSPs) [3,19]. The other ORFs encode at least four main structural proteins: the spike (S), membrane (M), envelope (E), and nucleocapsid (N) proteins [17,20,21].

Mpro, also known as 3-chymotrypsin-like protease (3CLpro), is a 33.8 kDa, three-domain cysteine protease, essential for proteolytic maturation and viral replication [9,18,22,23]. Mpro was found to be conserved among coronaviruses (CoVs), along with some common features of its substrates in different CoVs [18,24]. In addition to its vital role in the SARS-CoV-2 life cycle, the absence of any closely related human homologous Mpro makes it an ideal protein target for potential antiviral drugs, as its inhibitors are unlikely to be toxic in humans [25]. Furthermore, vaccines, as we have learned from previous viruses, can represent a selection pressure resulting in the evolution of novel resistant viral mutants, which again highlights SARS-CoV-2 Mpro as a good drug target, as it is less subject to such selection pressure caused by vaccines targeting the viral spike protein [26,27].

Enabled by the development of bioinformatics tools along with eased access to protein databases, virtual screening has proven to be a fundamental tool in drug design and drug repurposing research [28,29]. In the virtual screening approach, automated molecular docking tools are usually used to predict the best possible variant for binding one molecule to another, considering the best orientation with the best binding affinity [30]. These tools enable the screening of large numbers of candidates against a specific studied target, at a very low cost [31]. Virtual screening is a data-driven approach that can be either target-based, where a library of candidate ligands is docked against the target and analyzed, or ligand-based, where a similarity search or a machine learning strategy can be applied [32,33,34,35].

Since the beginning of the COVID-19 pandemic, a large number of studies around the world have used this approach to search for potential inhibitors of SARS-CoV-2 [22,36,37,38]. Joshi R.S. et al. used this approach in their study conducted in 2020 to scan over 7000 compounds from different origins against SARS-CoV-2 Mpro [39]. Another study conducted by Tallei E.T. et al. in 2020 used this approach to evaluate the potency of plant-derived bioactive compounds against Mpro, resulting in the identification of pectolinarin, hesperidin, nabiximols, rhoifolin, and epigallocatechin gallate as potential antiviral phytochemicals [40]. Research by Tahir Ul Qamar et al. also resulted in the identification of 5,7,3′,4′-Tetrahydroxy-2′-(3,3-dimethylallyl) isoflavone, amaranthin, licoleafol, calceolarioside B, and methyl rosmarinate as potential inhibitors of the target, using this approach [41]. Khaerunnisa S. et al. extended the list with kaempferol, quercetin, luteolin-7-glucoside, demetoxycurcumine, naringenin, apigenin-7-glucoside, oleuropein, catechin, curcumin, zingerol, gingerol, and allicin [42]. Essential oils have also shown their effectiveness against SARS-CoV-2 Mpro in silico [43,44,45]. Therefore, using this approach, multiple natural compounds have been identified as strong binders of SARS-CoV-2 Mpro, and some of them were also identified as multi-target inhibitors that could be applied in COVID-19 management approaches [36,37,38,39,40,41,42,43,44,45].

High-throughput screening (HTS) is a method for automated testing of thousands to millions of compounds for their biological activity against specific targets on model systems [46]. The development of robotics, laboratory equipment, laboratory methods, and software for the control of sample preparation, incubation, results detection, and data processing has allowed the HTS approach to be used to quickly search for lead compounds. Therefore, it is possible to quickly and inexpensively test large libraries of chemical compounds for their biochemical activity [47].

In practice, HTS is implemented in the form of a large number of miniature in vitro assays to identify molecules that can modulate the activity of a biological target. These reactions are run in 96-well, 386-well, or 1536-well plates [48]. Most often, the results of such biochemical analyses are obtained using various fluorescence detection methods [49], for example, direct measurement of fluorescence, fluorescence polarization, fluorescence resonance energy transfer (FRET), fluorescence quenching energy transfer (QFRET), or time-resolved fluorescence [46].

Since the early stage of COVID-19 pandemic, a large number of HTS assays have been developed worldwide to screen huge libraries of either previously approved drugs or potential inhibitors against SARS-CoV-2 [50,51,52,53]. Using this approach in their study to screen a library of 10,755 potential inhibitory compounds against SARS-CoV-2 Mpro, along with drugs previously approved for other viruses, Zhu W. et al. identified 23 potential inhibitors with different half-maximal inhibitory concentration values (IC₅₀), the efficacy of 7 of which was confirmed in a later cytopathic effect assay [54]. Given the high safety level required of laboratories studying and manipulating the live SARS-CoV-2 virus (BSL-3 laboratories), Zhang, Q.Y. et al. proposed a new HTS assay to enable potential antiviral testing in a BSL-2 research facility, where they constructed a reporter replicon of the virus using Renilla luciferase (Rluc) reporter gene and validated it later using hit natural compounds [52]. Froggatt H. M. et al. also developed a fluorescence-based HTS assay using a protein derived from green fluorescent protein (GFP) to serve as a target for SARS-CoV-2 Mpro, and hence a reporter for the enzyme’s inhibition and activation, enabling rapid screening of libraries and identification of lead compounds [55]. A further improvement of HTS can be achieved by combining it with the previous in silico approach to yield an ultra-high-throughput virtual screening approach, where huge libraries can be tested against multiple viral targets efficiently and rapidly [56]. Gorgulla C. et al. conducted a study to search for SARS-CoV-2 inhibitors using this large-scale HTS screening approach and were able to screen over one billion candidate molecules against 40 different target sites on 17 potential targets, both in the virus and the host [56]. Although the results were obtained from computational data and have not all been tested with experimental analyses yet, this filtration of candidates could narrow down research targets for later more detailed and efficient analyses.

Cell-based assays offer an advantage over virtual or biochemical screening assays, as they provide a whole physiological environment, reflecting the complexity of a living system rather than focusing on a specific isolated target and thereby enabling a more accurate evaluation of the biological activity and potential toxicity of screened compounds [57,58,59]. Due to practical considerations, it is important to develop and test drug compounds that exhibit inhibitory activity at various stages of the virus life cycle. Therefore, test systems have been developed to evaluate the effectiveness of inhibitors of entry, uncoating, replication, assembly (in which viral proteases are active), and maturation of viruses [60]. However, the whole variety of such systems can be reduced to two main mechanisms for their implementation: cytopathic and reporter mechanisms [61]. In the first case, the activity of antiviral agents is assessed by reducing the formation of plaques due to the accumulation of coloring or luminescent agents in living cells [62,63]. In the second, viruses and cells with report inserts are used, and the activity of inhibitors helps in reducing the expression of the reporter protein [64].

Since work with a live virus is accompanied by significant organizational restrictions, approaches have been developed for evaluating the effectiveness of antiviral agents that model various stages of the life cycle of viruses in cells of HeLa [65], Escherichia coli [66], and Saccharomyces cerevisiae [61]. Moreover, cell-based assays are nowadays increasingly integrated into HTS assays to accomplish rapid screens in a relevant physiological environment [58].

Several recent studies have used antiviral activity cell-based assays, either after or combined with the previously described approaches, to investigate previously approved drugs and herbal medicines for their potential inhibition potency against SARS-CoV-2 [9,67,68,69,70]. Applying this methodology followed by a further in vivo validation, Jan J.T. et al. screened a 3000-candidate library of both pharmaceuticals and herbal medicines to test their effectiveness against SARS-CoV-2 Mpro and RNA polymerase and proposed multiple herbal extracts as potential herbal inhibitors against the targeted viral enzymes [68]. Another recent study conducted by Qiao J. et al. also applied this approach to investigate 32 different inhibitors against SARS-CoV-2 Mpro, 6 of which were found to have a high inhibition potency and were used to select candidates for further in vivo investigation [71].

The phased use of these three approaches makes it possible to identify those that exhibit the targeted therapeutic activity from the variety of known plant secondary metabolites. Among these compounds, there may be those that have not previously exhibited such properties. Furthermore, compounds for which hypothetical activity is found can be quickly tested for their effectiveness on cell-free and then on cellular systems. Such a screening strategy has shown to be effective in the search for inhibitors of SARS-CoV-2 Mpro, which indicates its potential in the search for drugs against new pathogens [9].