Trends Sci. 2026; 23(2): 11438

SARS-CoV-2 is an infectious disease caused by a novel coronavirus that was first identified in December 2019 in Wuhan, China. Notably, its rapid global spread prompted the World Health Organization to declare it a global pandemic [1,2] . SARS-CoV-2 belongs to the family Coronaviridae and is characterized by a diameter of 60 - 140 nm; 9 - 12 nm spike proteins, which give it a solar corona-like appearance [3]. Vaccines are available and have significantly reduced morbidity and mortality. However, their long-term effectiveness is challenged by the frequent emergence of new viral variants, particularly those with mutations in the spike protein, a primary target for neutralizing antibodies [4]. Such mutations may compromise vaccine performance and highlight the need for effective antiviral therapies.

Although several drugs, such as remdesivir have been approved by Food and Drug Administration (FDA) or have received Emergency Use Authorizations (EUAs) such as lopinavir, ribavirin, ritonavir-boosted nirmatrelvir (paxlovid), chloroquine and ivermectin as antiparasitic agents, and the antibiotic azithromycin, which have been employed individually or in combination as analog therapies for the treatment of SARS-CoV-2, most of them are repurposed drugs originally developed for other diseases. Some of these drugs, also remains limited by issues such as high cost, limited availability in some regions, and administration requirements (e.g., intravenous remdesivir) [5,6]. The continuous emergence of viral variants further threatens the durability of these therapies. Therefore, discovering novel compounds with simpler structures, potential oral bioavailability, and affordable synthesis is essential, particularly in low-resource settings. Such efforts are aligned with the United Nations Sustainable Development Goals (SDGs), especially Goal 3 (Good Health and Well-Being) and Goal 10 (Reduced Inequalities), by promoting equitable access to effective and sustainable treatments [7].

To address these limitations, phenotypic screening has re-emerged as a valuable strategy. Phenotypic screening is an approach in drug development that focuses on observing and understanding phenotypic changes that occur at the cellular, tissue, or whole organism level in response to certain substances or chemical compounds [8]. Unlike target-based screening, which requires prior knowledge of specific molecular targets such as enzymes or receptors and may overlook compounds with broader or multi-target effects [9], phenotypic screening approach directly evaluates the biological effects of compounds in infected cells. This approach captures complex cellular interactions and increases the likelihood of identifying novel scaffolds with antiviral activity, even when their mechanisms of action are not fully understood. By starting from observed cellular responses, phenotypic screening accelerates the identification of active molecules while minimizing the risk of missing compounds with multi-target or indirect effects [10] . After screening candidate compounds on cells and microorganisms, potential compounds are further studied for their pharmacokinetic properties, such as absorption, distribution, metabolism, and excretion (ADME) [11]. Importantly, while phenotypic screening serves as the primary strategy, complementary target-based analyses such as molecular docking were employed to provide molecular insights [12]. This integration enables a better understanding of the observed phenotypic effects and offers preliminary explanations of the possible modes of action of the candidate compounds.

Heterocyclic compounds such as dihydropyrimidinones (DHPMs) [13,14], chalcones [15], and chromenes [16] represent promising scaffolds for this purpose. They are easily synthesized, structurally diverse, and they exhibit broad pharmacological activities including antiviral, antibacterial, antiparasitic, antifungal, anti-inflammatory, and anticancer properties [17].

DHPM and pyrimidine derivatives are recognized as nucleobase analogs with activity against viral polymerases and proteases [18]. Chromenes, structurally related to flavonoids, have been linked to antiviral and anti-inflammatory effects [19], while chalcones, with their reactive enone system, display broad-spectrum, multi-target antiviral and immunomodulatory potential [20]. Both chalcone and chromene derivatives have shown consistent activity against RNA viruses, including HIV, HCV, and influenza. Their versatility makes them attractive candidates for drug discovery programs targeting emerging viral infections.

However, despite these interesting findings, studies exploring their potential against SARS-CoV-2 remain limited. Therefore, this study aimed to conduct a preliminary evaluation of chalcone, chromene, and DHPM derivatives against SARS-CoV-2. To achieve this, a phenotypic screening approach was employed using a local isolated viral strain, assessing cytotoxicity and antiviral activity. This was followed by in silico ADME profiling to predict pharmacokinetics and molecular docking, including grid score and docking pose/pharmacophore score, to explore possible mechanisms of action targeting the main protease enzyme of SARS-CoV-2. As an early-stage investigation, this comprehensive approach is expected to facilitate the identification of promising lead compounds for further development as anti-SARS-CoV-2 therapeutics.

Materials and methods

Materials

All chemical reagents, including 100% dimethyl sulfoxide (DMSO), [3-(4,5-dimethylthiazol-2yl)-5(3-carboxymethoxyphenyl)-2-(4-sulfophenyl0-2H-terazolium] (MTT), phosphate buffer saline (PBS), and 1% penicillin-streptomycin, were obtained from Sigma-Aldrich, USA. Sterilization was performed using 0.22 µm membrane filters (OMNIPORE, Millipore). Cell culture was performed using Dulbecco’s Modified Eagle Medium (DMEM) with glucose (Biowest), supplemented with 5% Fetal Bovine Serum (FBS) from Biowest, and Trypsin TPCK (Worthington Biochemical Corporation, USA). RNA extraction was performed using the RNeasy Kit (QIAGEN, USA), and qRT-PCR was conducted using the STANDARD™ M nCoV Real-Time Detection Kit (SD Biosensor Inc., Korea) [21]. Ten synthetic compounds (4 DHPMs, 2 chromenes, and 4 chalcones) were sourced from BIOME, Universitas Airlangga. Vero ATCC CCL-81 ( https://www.atcc.org/products/ccl-81 ) cells were provided by the Center of Infectious Diseases, Kobe University, Japan, and the SARS-CoV-2 virus isolate (B1.470; GISAID ID: EPI_ISL_529965, Clade GH) [22,23] was obtained from a patient. Reference drug structures were obtained from an online database https://pubchem. ncbi.nlm.nih.gov. The enzyme used for the in-silico study was SARS-CoV-2 Main Protease (Mpro) complexed with the non-covalent ligand N-(4-tert-butylphenyl)-N-[(1R)-2 (cyclohexylamino)-2-oxo-1- (pyridin-3-yl)ethyl]-1H-imidazole-4-carboxamide (X77) (PDB ID: 6w63), was retrieved from the RCSB Protein Data Bank (website http://www.rcsb.org/pdb).

Instrumentation

MTT assay was conducted using a Spectroreader Multidetection Microplate Reader Multiskan SkyHigh Photometer (Thermo Fisher Scientific). Virus amplification was performed via qRT-PCR using the Applied Biosystems 7500 Fast system and its accompanying software (Applied Biosystems). GraphPad Prism 8.3.0 software was used to analyze the cytotoxicity and dose-response data. The of ADME properties of the candidate ligands were predicted using the SwissADME tool (https://www.swissadme.ch). Toxicology evaluation was conducted using the Toxicity Estimation Software Tool (T.E.S.T); https://www.epa.gov/comptox-tools/toxicity-estimation-software-tool-test to assess Ames mutagenicity (M) and LD ₅₀ (mg/kg). Molecular docking analysis was performed using software on 2 operating systems: Windows 10 and Linux. On the Windows system, ChemDraw Professional 15.1, Chimera 1.16, PuTTY (64-bit), WinSCP, and Discovery Studio BIOVIA 2021 2021 v21.1.0.2098 were used to generate 2-dimensional (2D) visual representations of ligand-receptor interactions. On the Linux system, the DOCK 6 program was employed for molecular docking, and Gaussian 16 was utilized for computational calculations.

Cell preparation and virus propagation

Vero cells were cultured in DMEM supplemented with 5% FBS and 1% penicillin-streptomycin and passaged at cell confluency >80%. SARS-CoV-2 was cultured on Vero cells in the presence of 2.5 µg/mL of TPCK-treated trypsin. After centrifugation, the viral supernatant was further propagated on Vero cells to produce a working viral stock. Vero cells used in this study were at passage number 55. Specifically, the stock was prepared by diluting the virus supernatant with 0.2% Bovine Serum Albumin (BSA) in Tris-Glycine-Saline (TGS) in a 2:1 ratio. Notably, viral titer was determined using the 50% tissue culture infectious dose (TCID50) culture method. Briefly, Vero cells were seeded in a 96-well plate at a density of 1×10 ⁴ cells/well for 24 h, followed by dilution at 10 - 1 to 10 - 10. Finally, TCID50 was determined based on the cytopathic effect of the cell infection as well as with qRT-PCR and calculated using the Reed and Munch method [24].

Cell viability assay

Candidate compounds were dissolved in 100% DMSO to prepare 1 M stock solutions. Thereafter, these solutions were filtered through a 0.22 μm membrane under aseptic conditions [25]. Dilutions were prepared in complete DMEM to achieve the final test concentrations.

Cell viability was assessed using the MTT assay (Sigma-Aldrich) as described by Permanasari et al . [26]. Briefly, Vero cells were seeded in 96-well microplates (1×10 ⁴ cells/well) containing 100 μL of DMEM supplemented with 5% FBS and 1% penicillin-streptomycin at 37 °C for 24 h under a 5% CO ₂ atmosphere.

Thereafter, Vero cells were treated with the test compounds at dilutions of 1600, 800, 400, 200, 100, and 50 µM in triplicate and incubated for 48 h. Finally, the absorbance was measured at 560 and 750 nm. Cell viability was determined using the following equation [27,28]: % Viable cells = (Treatment Absorbance – Control Medium Absorbance)/(Negative Control Absorbance – Control Medium Absorbance)×100%. Results were obtained using a nonlinear regression equation (analysis 4-parameter logistic model) using GraphPad Prism 8.3.0 software. The results are expressed as the 50% Cytotoxic Dose (CC ₅₀ ).

In vitro anti SARS-CoV-2 assay

Vero cells were seeded in a 96-well plate at a density of 1×10 ⁴ cells/well and incubated for 24 h at 37 °C in a 5% CO₂ atmosphere. After removing the culture medium, the cells were infected with 100 TCID₅₀/mL of SARS-CoV-2 virus diluted in DMEM containing 2.5 µg/mL TPCK-treated trypsin, followed by gentle shaking and 1 h incubation [29]. The viral suspension was then removed, and serial concentrations of the test compounds (10, 20, 40, and 80 µM) were added to the infected cells, with each treatment performed in duplicate. The plates were incubated for 72 h under the same conditions, and observations were recorded at 24, 48, and 72 h. Infected Vero cells treated with 0.1% DMSO served as negative controls to establish the baseline level of viral replication. No positive control (e.g., known antiviral drug) was included, as this study was designed as a preliminary phenotypic screening to identify novel compounds with measurable antiviral activities, rather than to directly compare their effects with established agents.

After incubation, RNA was extracted from the supernatants using qRT-PCR. Antiviral activity was quantified using the following formula where A 0 represents the Average RT-qPCR copy number without a sample, and A 1 represents the Average RT-qPCR copy number with the sample: % Inhibition = .The IC₅₀ value was determined using nonlinear regression analysis (4-parameter logistic model) with GraphPad Prism 8.3.0. One-way ANOVA with Tukey’s post hoc test was used to compare IC₅₀ values with statistical significance was set at p < 0.05 [30]. The selectivity index (SI) values were determined by calculating the ratio of CC ₅₀ to IC _50, and the standard deviation of the SI was calculated using the standard error propagation method. Compounds with an SI >10 were considered potential drug candidates.

RNA extraction and amplification of SARS-CoV-2

SARS-CoV 2 RNA was extracted from fluid samples using the RNeasy kit (Qiagen, Valencia, Calif., USA). Eighty microliters (80 μL) of the culture supernatant were mixed with 320 µL of AVL lysis buffer (Qiagen, Valencia, Calif., USA). After adding 320 µL of 96% ethanol, the samples were filtered using a spin column RNeasy kit to extract the RNA. The extracted RNA was eluted in 60 µL of nuclease-free water.

qRT-PCR was performed to amplify the ORF1ab (RdRp) and E genes using the STANDARD™ M nCoV Real-Time Detection kit (SD Biosensor Inc., Suwon, South Korea). A positive result was reported if ORF1ab or both genes had Ct values within the valid range (<38), and samples with endpoints above the minimum threshold were considered presumptive positives. In contrast, the sample was considered negative if neither target met these criteria. Presumptive positive results were treated as positive owing to their strong indication of SARS-CoV-2 presence [31].

Experimental design and bias control

As this was an initial screening study, formal randomization was not applied . To reduce potential bias, all compounds were anonymized using coded identifiers (e.g., S-1, S-10, C-3, and C-5), and distributed in a balanced, non-systematic order across microplates to minimize the edge effects and positional bias. Cytotoxicity (MTT assay) and antiviral activity (qRT-PCR) were independently assessed by 2 researchers using instrument-based measurements. Blinding of compound identities further ensured objectivity.

In silico analysis

ADME predictions

Pharmacokinetic analysis of drug similarity properties, including drug-likeness, was performed using the SwissADME program [32,33]. Generally, drug candidate selection is based on ADME properties, according to Lipinski’s rule of 5 and the Ghose, Veber, Egan, and Muegge rules [34,35]. Additionally , the Toxicity Estimation Software Tool (T.E.S.T) was used to predict the toxicity and mutagenicity of the compounds, based on various parameters, including LD ₅₀ and the results of the Ames test for mutagenicity. LD ₅₀ provides an indication of the toxicity of a substance, with a lower LD ₅₀ value indicating higher toxicity and a mutagenicity index ≤ 1 indicating no significant increase in mutagenicity. Notably, the closer the total score is to 1, the safer the compound [36,37].

Molecular docking

To prepare SARS-CoV-2 main protease receptor (PDB ID: 6w63) , water molecules and non-complex ions were removed, occupancy selection was performed, and charges were added, using ff14SB for the protein and AM1-BCC for non-protein parts [38]. The protein, receptor surface grid and spheres were generated using the molecular docking software Dock-6 [39] and Chimera software. Both programs operate through server usage, necessitating the setup of communication software, such as WinSCP and PuTTY.

Reference drug compounds were sourced from PubChem, and ligand structures in SMILES format were converted into 3D structures using the CACTUS server (https://cactus.nci.nih.gov). Subsequently, the structures were optimized using the PM7 method in Gaussian16, and charges were assigned using Antechamber. Finally, the ligands, receptors, and complexes were saved in the MOL2 format [40].

Docking was performed with a flexible docking approach using DOCK6, and the protocol was validated by re-docking the co-crystallized ligand (X77) onto the receptor. Successful docking validation was confirmed if the ligand’s pose matched its original conformation with an RMSD ≤ 2 Å, supported by consistent interaction profiles. Here, ‘original conformation’ refers to the co-crystallized orientation of X77 within the active site of SARS-CoV-2 Mpro (PDB ID: 6w63). Finally, candidate ligands were docked onto the main protease receptor, with selection based on grid scores and generated poses [41].

Results and discussion

In this study, we investigated the antiviral effects of 10 compounds derived from DHPMs, chromene (benzopyrene), and chalcone derivatives will be observed as candidate anti-COVID-19 drugs. DHPMs and Pyrimidinones are heterocyclic compounds synthesized through classic multi-component reactions, such as the Biginelli reaction [42] .

Table 1 Drug candidate compounds.

In Table 1 , compound-2 (S-2), compound-3 (S-3), compound-4 (S-4), compound-12 (S-12) are DHPM and pyrimidinones, whereas compound-10 (S-10) and compound-11 (S-11) are chromene. Chromenes also known as benzopyrans, are important compounds that are synthesized for commercial use. Chromenes are characterized by the presence of an oxygen atom in the pyran ring. Chromenes exist in various forms, such as simple chromenes (2H-chromene) and fused chromenes. Recent advances have demonstrated various eco-friendly and efficient strategies for their synthesis, including multicomponent reactions, visible-light photocatalysis, and 1-pot protocols using green catalysts and solvents [43,16].

Chalcones are a group of compounds categorized as flavonoids due to their structural scaffold, which is C6-C3-C6. Chalcone, derivatives can be obtained from natural sources or as synthetic products [44,45]. As shown in Table 1 , compound-1 (C-1), is an amino chalcone, whereas the other chalcones, including compound-5 (C-5), compound-6 (C-6), and compound-7 (C-7). For Compound C-5, the hybrid combination was achieved by combining active pharmacophores (coumarin and thiazole), whereas the others originated from chalcone thiourea (C-6) and chalcone thiourea executed with thiophene (C-7).

Cytotoxicity of the derivatives

A cell viability assay was performed to assess the cytotoxicity of the 10 compounds, and the results were expressed as CC ₅₀ . As shown in Figure 1 , there was a concentration-related reduction in Vero cell viability after treatment with the 10 candidate compounds at concentrations of 50, 100, 200, 400, 800, and 1600 µM

Figure 1 Cytotoxicity of chalcone (C-1, C-5, C-6, C-7), chromene (S-10, S-11), and dihydropyrimidinone (S-2, S-3, S-4, S-12) derivatives in Vero cells at concentrations ranging from 50 to 1600 µM. Data are presented as mean ± SD (each concentration was repeated in triplicate). Color and symbol codes: C-1 (blue-round), S-2 (red-square), S-3 (red-triangle), S-4 (red-invertedtriangle), C-5 (blue-diamond), C-6 (blue-square), C-7 (blue-triangle), S-10 (green-*), S-11 (green-star), and S-12 (red-diamond).

Among them, S-10, S-11 and S-12 demonstrated low cytotoxicity, C-5, S-4, S-2 and S-3 showed moderate cytotoxicity and C-1, C-6 and C-7, exhibited high cytotoxicity.

The CC ₅₀ values of the compounds are listed in Table 2 . These values were calculated from dose-response curves using nonlinear regression analysis (4-parameter logistic model) in GraphPad Prism version 8.3.0. Cell viability percentages were plotted against compound concentrations, and the CC₅₀ values were derived from the best-fit curves.

Table 2 Cytotoxicity of chalcone, chromene, and DHPM derivatives in Vero cells.

Notes: *>200 µM Moderate Cytotoxicity; 20 - 100 µM Strong Cytotoxicity; 1 - 20 μM Very Strong Cytotoxicity (Indrayanto et al . [46]) .

Based on a previously established classification system [46], compounds with CC₅₀ values of 1 - 20 µM are considered extremely cytotoxic; those with values between 20 and 100 µM are categorized as strongly cytotoxic; and compounds with CC₅₀ values ≥200 µM are classified as moderately cytotoxic. Based on this classification, 5 compounds (C-1, C-6, C-7, S-2, and S-3) with CC₅₀ values close to 100 µM were categorized as having strong cytotoxicity and were therefore not considered for further antiviral evaluation. Consequently, 5 compounds S-4, C-5, S-10, S-11, and S-12 were selected for further analysis.

In vitro anti-SARS-CoV-2 viral activity

Five compounds S-4, C-5, S-10, S-11, and S-12 were selected based on preliminary cytotoxicity results for further evaluation of their anti-SARS-CoV-2 activity. The IC₅₀ values were determined from dose-response curves using nonlinear regression (4-parameter logistic model) in GraphPad Prism 8.3.0. as described in Table 3 .

Table 3 Antiviral activities of the compounds .

Notes: *a) Criteria of IC50: 1 - 20 µM good activity, 20 - 100 µM moderate activity (Indrayanto et al. [46]). *b) Criteria of Selectivity Index >10 very safe, 10 - 3 moderate (Indrayanto et al . [46]).

Figure 2 shows the IC ₅₀ values of the compounds against SARS-CoV-2. According to the established classification criteria Batista et al ., 2009 in Indrayanto et al . [46] compounds with IC₅₀ values < 20 µM are considered to have strong antiviral activity. In this study, C-5 (5.62 ± 0.24 µM), S-12 (6.19 ± 0.41 µM), and S-10 (8.52 ± 0.28 µM) demonstrated the highest potency, whereas S-4 (25.21 ± 5.38 µM) was moderately active.

Figure 2 Dose-response curves of selected chalcone (C-5), chromene (S-10, S-11), and dihydropyrimidinone (S-4, S-12) derivatives showing antiviral activity against SARS-CoV-2 in Vero cells. The tested concentrations were 10, 20, 40, and 80 µM. Color codes: S-4 (blue), C-5 (red), S-11 (green), S-10 (purple), and S-12 (orange). Data are shown as mean ± SD. Each concentration was tested in triplicate.

Statistical analysis using one-way ANOVA followed by Tukey’s post hoc test ( Table 3 ) revealed significant differences among several compound pairs. Notably, compounds C-5, S-10, and S-12, which exhibited the lowest IC₅₀ values, were statistically different from the other tested compounds ( p < 0.05). Although there were no significant differences among C-5, S-10, and S-12 themselves, each showed significantly lower IC₅₀ values compared to the remaining compounds.

Table 3 Pairwise IC₅₀ comparison (one-way ANOVA followed by Tukey’s post hoc test).

Notes: * p < 0.05, ** p < 0.01, *** p < 0.001.

The selectivity index (SI), defined as the ratio of CC₅₀ to IC ₅₀, indicates the margin of safety for each compound. It was used to evaluate the therapeutic potential. Based on the data in Figure 3 , all tested compounds had SI ≥ 10, with S-12 (SI = 309.7), C-5 (SI = 161.3), and S-10 (SI = 158.8) showing the most favorable profiles for further evaluation. In contrast, the moderate SI values for S-4 and S-11 (17.1 and 111.9, respectively) suggest that structural modifications are necessary to improve their efficacy and safety profiles.

Figure 3 IC ₅₀ , CC ₅₀ , and Selectivity Index (SI) values (mean ± SD) of selected compounds. The IC ₅₀ values (red), represent the concentration required to inhibit 50% of SARS-CoV-2 viral activity in Vero cells, while the CC ₅₀ values (blue), represent the concentration at which 50% of the cells exhibit cytotoxicity. The selectivity index (SI) is the ratio of CC ₅₀ to IC ₅₀ (green), indicating the margin of safety for each compound.

Although direct comparisons with FDA-approved commercial antivirals such as remdesivir were not performed in this study, the IC₅₀ values of C-5, S-10, and S-12 were found to fall within the low micromolar range. For contextual reference, previous studies have reported IC₅₀ values of remdesivir between 0.77 and 1.65 μM in Vero E6 and human lung cells [47,48]. These comparisons are presented merely as background information and should not be interpreted as evidence of equivalent efficacy. Instead, the present findings highlight the preliminary antiviral potential of the tested compounds, indicating the need for further investigation and direct comparative studies with standard drugs such as remdesivir to clarify their relative effectiveness.

ADME characteristics of the compounds

To support further evaluation of the selected compounds, an in silico analysis of absorption, distribution, metabolism, and excretion (ADME) properties was conducted. This analysis aimed to estimate drug-likeness and potential pharmacokinetic behavior by applying Lipinski’s rule of 5, as well as the Ghose and Veber criteria. The e valuated parameters included molecular weight (MW), number of hydrogen bond donors and acceptors, rotatable bonds, molecular refractivity (MR), topological polar surface area (TPSA), predicted octanol-water partition coefficient (WLOGP), and blood-brain barrier (BBB) permeability [49].

Table 5 ADME prediction analysis.

Notes: MW = Molecular weight, MR = Molar refractivity, TPSA = Polar surface area, Log P = Water partition coefficient, BBB = Blood-Brain Barrier permeability, LD ₅₀ = 50% lethal-dose, Mut = Ames’s mutagenicity, SS ADME = ADME Selection Score.

According to Table 5 , all 5 compounds that passed in vitro screening (C-5, S-4, S-10, S-11, and S-12) fulfilled the general molecular weight requirement (MW < 500), which correlates with enhanced cellular permeability and oral bioavailability. Additionally, all compounds had TPSA values below 120 Ų, indicating favorable potential for passive gastrointestinal absorption. The WLOGP values of most compounds also fell within the optimal range of –0.4 to 5.6, suggesting a balanced lipophilicity. However, compound C- 5 exhibited a WLOGP value exceeding this threshold, indicating a risk of low solubility. Poor solubility can negatively impact drug absorption and bioavailability, especially for orally administered drugs [37,34].

Regarding BBB permeability, none of the compounds were predicted to cross the blood-brain barrier, which may be advantageous in avoiding central nervous system (CNS) side effects for drugs targeting respiratory viral infections. Toxicity risk was further assessed using the Selection Score (SS), which integrates ADME parameters and predicted toxicity. Compounds S-4, S-10, S-11, and S-12 demonstrated SS values closer to 1, indicating more favorable predicted safety and pharmacokinetic profiles than the other compounds [50,51].

Overall, while in silico ADME predictions suggested acceptable pharmacokinetic characteristics for 4 of the 5 selected compounds, it is important to note that these findings remain predictive. Experimental validation through solubility testing, permeability assays, and pharmacokinetic studies is necessary to confirm their drug-like behavior in vivo is necessary.

Therefore, S-10 and S-12 have high selectivity index values, favorable in silico ADMET profiles, and structurally simpler scaffolds, making them attractive candidates for further optimization. Unlike Remdesivir, which requires intravenous administration and intracellular activation, the tested compounds (S-10 and S-12) show potential for oral bioavailability and cost-effective synthesis, offering significant advantages for broad antiviral development, especially in low-resource settings.

The molecular docking analysis

The SARS-CoV-2 main protease (Mpro) is a cysteine protease derived from a polyprotein that forms a homodimer with catalytic residues His-41 and Cys-145 at the active site. Cys-145 functions as a nucleophile, while His-41 assists in proton transfer, making both essential for enzymatic cleavage [52,53]. The co-crystallized ligand X77, used as a docking reference, engages in van der Waals interactions and hydrogen bonds with key residues such as Ser-144, His-163, and Glu-166, along with a salt bridge with His-163 [54]. As summarized in Table 6 , the docking results indicated that 4 candidate compounds (S-4, S-10, S-11, and S-12) interacted with key catalytic residues (Cys-145 and His-41). These interactions include hydrogen bonding, hydrophobic contacts , π – π stacking, and electrostatic interactions, all of which are essential for inhibitory activity.

Importantly, the observed interactions with His-41 and Cys-145 are consistent with those reported for known SARS-CoV-2 protease inhibitor, suggesting potential biological relevance in enzyme inhibition [55,56]. Nevertheless, docking results represent predictive binding affinity and should be interpreted cautiously, as this approach does not account for protein flexibility, solvent effects, or pharmacokinetic properties. Therefore, the molecular docking data in this study are intended to complement and support the preliminary in vitro findings, rather than to provide definitive evidence of antiviral efficacy [57,58].

Tabel 6 Docking results of the compounds on Mpro .

Molecular docking visualization

Chromene derivatives

Docking analysis revealed that S-10 and S-11 as shown in Figure 4(A) (yellow and red) closely resembled the orientation of the reference inhibitor X77, the original ligand of the main protease (PDB ID: 6W63, Figure 4(B) ; magenta) . Both compounds formed interactions with the catalytic residues His-41 and Cys-145 . For S-10, the structure includes amino, hydroxyl, and phenyl groups, which facilitate hydrogen bonding and hydrophobic interactions. The hydroxyl and amino groups enhanced hydrogen bonding with the Mpro catalytic dyad, while the phenyl group established hydrophobic contacts within the S1 pocket of the Mpro active site. These combined interactions stabilized the binding orientation and improved antiviral potency [59].

Meanwhile, S-11, which contained diamino, phenyl, and carbonitrile groups, adopted a less favourable binding mode. The high polarity of diamino substituent increased hydrophilicity, thereby limiting hydrogen bond formation at the catalytic site and reducing membrane permeability, which in turn decreased intracellular accumulation. These molecular features are consistent with the biological data, where S-10 exhibited stronger antiviral activity (IC₅₀ = 8.52 μM, SI = 158.8) compared to S-11 (IC₅₀ = 19.48 μM, SI = 111.9).

DHPM derivatives

Docking analysis revealed that S-12 ( Figure 4(A) ; light blue color) closely resembled the binding orientation of the reference inhibitor X77, the original ligand of Mpro (PDB ID: 6W63, Figure 4(B) ; magenta). The compound exhibited strong interactions within the active site by forming hydrogen bonds with the catalytic residues His-41 and Cys-145, π – π stacking with His-41, and a salt bridge with His-163. These multiple interactions stabilized the ligand within the pocket. In addition, the carbonyl and phenyl groups improved binding stability, while the carboxylate group enhanced aqueous solubility and electrostatic stabilization, further supporting ligand bioavailability and target engagement.

In comparison, S-4 ( Figure 4(A) ; purple color) adopted a less favorable orientation relative to X77 ( Figure 4(B) ). It interacted with His-41 primarily through van der Waals forces, which were insufficient to secure strong binding. Structurally, S-4 contained thiophene, methyl, and thioxo groups that increased lipophilicity but provided limited opportunities for hydrogen bonding with the catalytic residues, leading to weaker stabilization in the active site [59].

These docking observations are in line with the biological results, where S-12 demonstrated the strongest antiviral effect among the DHPM derivatives (IC₅₀ = 6.187 μM, SI = 309.7), while S-4 showed weaker activity (IC₅₀ = 25.21 μM, SI = 17.1). Overall, these results emphasize that polar functional groups capable of hydrogen bonding (e.g., hydroxyl, amino, carboxyl) are more favorable than lipophilic substituents in achieving potent inhibition of SARS-CoV-2 Mpro

.

Figure 4 Molecular docking visualization of selected compounds with SARS-CoV-2 main protease (Mpro, PDB ID: 6W63). (A) Superimposed docking poses of candidate compounds: S-10 (yellow), S-11 (red), S-12 (light blue), and S-4 (purple). ( B) Binding interaction of the co-crystallized ligand X77 (magenta) at the Mpro active site (visualized using the Biovia Discovery Studio).

Docking simulations in this study included the native ligand X77, FDA-approved antivirals (e.g., Remdesivir, Ritonavir, Lopinavir, Nirmatrelvir), and the tested compounds (S-4, S-10, S-11, S-12). As shown in Table 6 , the reference inhibitors consistently interacted with His-41 and Cys-145, confirming the catalytic dyad as a critical binding region. Among the candidate compounds, S-12 displayed the most favorable docking score and binding pose, showing interaction patterns comparable to X77 and several reference antivirals [60,61] . S-10 also engaged in hydrogen bonding and hydrophobic contacts with the catalytic residues, whereas S-4 and S-11 showed weaker scores and less optimal orientations. These findings suggest that S-12 and S-10 may possess favorable binding modes at the Mpro active site; however, it should be emphasized that docking provides only predictive interaction patterns [62] and does not account for protein flexibility or solvent dynamics, so further validation is required.

Conclusions

This study identified 2 lead compounds, S-12 and S-10, from DHPM, chromene, and chalcone scaffolds, which exhibited potent in vitro antiviral activity against a locally isolated SARS-CoV-2 strain, with low cytotoxicity, high selectivity index (>100), and favorable drug-like properties. In silico analyses further supported their oral bioavailability and the absence of major toxicity alerts. Although, their IC₅₀ values were higher than those of approved antivirals such as Remdesivir, based on literature reports, but their simpler structures, lower cytotoxicity, and synthetic accessibility suggest practical advantages as early-stage candidates. A key limitation, however, is the absence of direct in vitro comparison with clinically used antivirals, which would provide stronger context for potency evaluation. Molecular docking revealed stable interactions with the catalytic residues His-41 and Cys-145 of Mpro, supporting their relevance as potential protease inhibitors, though such predictions remain preliminary. Overall, these findings highlight S-12 and S-10 as promising scaffolds that indicate the need for further in vivo validation, pharmacokinetic studies, structural optimization, and direct benchmarking against reference antivirals to advance their development as region-specific anti-SARS-CoV-2 agents.

Acknowledgements

This Research is financially supported by Indonesian Ministry of Higher Education through Research Funds (DRTPM) 2023 in platform Doctoral Research Grant for 1 year study based on No. 114/E5/PG.02.00.PL/2023 ; UN3.LPPM/PT.01.03/2023 and fully support of BIOME as well as ITD Universitas Airlangga.

Declaration of generative AI in scientific writing

Generative AI tools (e.g., ChatGPT by OpenAI, Grammarly) were used exclusively to enhance the clarity, grammar, and readability of the manuscript. All scientific content, including study design, data analysis, interpretation, and conclusions, was entirely developed and verified by the authors. The use of AI was conducted under full human oversight, and the authors assume complete responsibility for the integrity and accuracy of the work. No AI tool is listed as an author or co-author.

Credit author statement

Dinar Adriaty: Writing - Original Draft prepara­tion, Conceptualization, Methodology, Investigation, Formal Analysis, Visualization, Software; Hery Suwito: Resources, Conceptualization; Data curation, Writing - Reviewing and Editing. Ni Nyoman Tri Pus­paningsih: Resources; Funding acquisition, Writing - Reviewing and Editing; Adita Ayu Permanasari: Su­pervision, Validation; Aldise Mareta Nastri: Supervision, Validation; Kazufumi Shimizui: Resources, Writing - Reviewing and Editing.

References

[1] H Zhu, L Wei and P Niu. The novel coronavirus outbreak in Wuhan, China. Global Health Research and Policy 2020; 5(1) , 6.

[2] B Hu, H Guo, P Zhou and ZL Shi. Characteristics of SARS-CoV-2 and COVID-19. Nature Reviews Microbiology 2021; 19(3) , 141154.

[3] A Gori, F Fama and C Genovese. Coronavirus disease 19 (COVID-19) . In : MCB Raviglione, F Tediosi, S Villa, N Casamitjana and A Plasència (Eds.). Global Health Essentials. Springer, Cham, Switzerland, 2023, p. 137-142.

[4] I Polatoğlu, T Oncu-Oner, I Dalman and S Ozdogan. COVID-19 in early 2023: Structure, replication mechanism, variants of SARS-CoV-2, diagnostic tests, and vaccine & drug development studies. MedComm 2023; 4(2) , e228.

[5] Purwati, A Miatmoko, Nasronudin, E Hendrianto, D Karsari, A Dinaryanti, N Ertanti, IS Ihsan, DS Purnama, TP Asmarawati, E Marfiani, Yulistiani, AN Rosyid, PA Wulaningrum, HW Setiawan, I Siswanto and NNT Puspaningsih. An in vitro study of dual drug combinations of anti-viral agents, antibiotics, and/or hydroxychloroquine against the SARS-CoV-2 virus isolated from hospitalized patients in Surabaya, Indonesia. PLoS One 2021; 16(6) , e0252302.

[6] H Zhou, WJ Ni, W Huang, Z Wang, M Cai and YC Sun. Advances in pathogenesis, progression, potential targets and targeted therapeutic strategies in SARS-CoV-2-induced COVID-19. Frontiers in Immunology 2022; 13 , 834942.

[7] JD Rigg. The Sustainable Development Goals (SDGS) . In : The companion to development studies. Routledge, Oxfordshire, 2024, p. 253-257.

[8] EL Berg. The future of phenotypic drug discovery. Cell Chemical Biology 2021; 28(3) , 424-430.

[9] F Vincent, A Nueda, J Lee, M Schenone, M Prunotto and M Mercola. Phenotypic drug discovery: Recent successes, lessons learned and new directions. Nature Reviews Drug Discovery 2023; 21(12) , 899-914.

[10] S Wang, Z Wang, L Fang, Y Lv and G Du. Advances of the target-based and phenotypic screenings and strategies in drug discovery. Internatioal Journal of Drug Discovery and Pharmacology 2022; 1(1) , 5-11.

[11] SE Rhabori, AE Aissouq, S Chtita and F Khalil. QSAR, molecular docking and ADMET studies of quinoline, isoquinoline and quinazoline derivatives against Plasmodium falciparum malaria. Structural Chemistry 2023; 34(2) , 585-603.

[12] AM Andrianov, YV Kornoushenko, AD Karpenko, IP Bosko and AV Tuzikov. Computational discovery of small drug-like compounds as potential inhibitors of SARS-CoV-2 main protease. Journal of Biomolecular Structure and Dynamics 2021; 39(15) , 5779-5791.

[13] I Amalina, NNT Puspaningsih and H Suwito. In silico analysis of pyrimidine derivatives as potential antibacterial agents. In : Proceedings of the International Conference on Advanced Technology and Multidiscipline, Surabaya, Indonesia. 2023, p. 50004.

[14] LHS Matos, FT Masson, LA Simeoni and M Homem-de-Mello. Biological activity of dihydropyrimidinone (DHPM) derivatives: A systematic review. European Journal of Medical Chemistry 2018; 143 , 1779-1789.

[15] AN Kristanti, H Suwito, NS Aminah, KU Haq, HD Hardiyanti, H Anggraeni, N Faiza, RS Anto and S Muharromah. Synthesis of some chalcone derivatives, in vitro and in silico toxicity evaluation. Rasayan Journal of Chemistry 2020; 13(1) , 654-662.

[16] A Chaudhary, K Singh, N Verma, S Kumar, D Kumar and PP Sharma. Chromenes - A novel class of heterocyclic compounds: Recent advancements and future directions. Mini Reviews in Medicinal Chemistry 2022; 22(21) , 2736-2751.

[17] V Raj and J Lee. 2H/4H-chromenes - A versatile biologically attractive scaffold. Frontiers in Chemistry 2020; 8 , 623.

[18] S Khasimbi, F Ali, K Manda, A Sharma, G Chauhan and S Wakode. Dihydropyrimidinones scaffold as a promising nucleus for synthetic profile and various therapeutic targets: A review. Current Organic Synthesis 2020; 18(3) , 270-293.

[19] BT Dharmapalan, R Biswas, S Sankaran, B Venkidasamy, M Thiruvengadam, G George, M Rebezov, G Zengin, M Gallo, D Montesano, D Naviglio and MA Shariati. Inhibitory potential of chromene derivatives on structural and non-structural proteins of dengue virus. Viruses 2022; 14(12) , 2656.

[20] SM Villa, J Heckman and D Bandyopadhyay. Medicinally privileged natural chalcones: Abundance, mechanisms of action, and clinical trials. International Journal of Molecular Sciences 2024; 25(17) , 9623.

[21] SY Ham, H Jeong, J Jung, ES Kim, KU Park, HB Kim, JS Park and KH Song. Performance of STANDARD ^TM M10 SARS-CoV-2 assay for the diagnosis of COVID-19 from a nasopharyngeal swab. Infection and Chemotherapy 2022; 54(2) , 360-363.

[22] Y Shu and J McCauley. GISAID: Global initiative on sharing all influenza data – from vision to reality. Euro Surveillance 2017; 22(13) , 30494.

[23] K Rahardjo, AM Nastri, JR Dewantari, RR Prasetya, J Wahyuhadi, G Soegiarto, L Wulandari, RA Setyoningrum, R Yudhawati, YK Shimizu, M Nishimura, Y Mori, Soetjipto, K Shimizu and MI Lusida . GISAID: EpiCoV database . 2020. https://www.epicov.org/epi3/frontend#3280a8 .

[24] M Brandolini, F Taddei, MM Marino, L Grumiro, A Scalcione, ME Turba, F Gentilini, M Fantini, S Zannoli, G Dirani and V Sambri. Correlating qRT-PCR, dPCR and viral titration for the identification and quantification of SARS-CoV-2: A new approach for infection management. Viruses 2021; 13(6) , 1022.

[25] G McDonnell and DM Hansen. Block’s disinfection, sterilization and preservation . 6 ^th eds. Lippincott Williams and Wilkins, Philadelphia, 2020.

[26] AA Permanasari, C Aoki-Utsubo, TS Wahyuni, L Tumewu, M Adianti, A Widyawaruyanti, H Hotta and AF Hafid. An in vitro study of an Artocarpus heterophyllus substance as a hepatitis C antiviral and its combination with current anti-HCV drugs. BMC Complementary Medicine and Therapies 2021; 21(1) , 260.

[27] R Ramadhan, K Ul-Haq, P Phuwapraisirisan, FA Puspitasari and H Suwito. α-glucosidase inhibitory activities and antioxidant properties of depsidone from Garcinia parvifolia . Rasayan Journal of Chemistry 2023; 16(3) , 1811-1817.

[28] H Suwito, R Ramadhan, A Abdulloh and KU Haq. Synthesis of dihydrotetrazolopyrimidine derivatives as anticancer agents and inhibitor of α-glucosidase. Rasayan Journal of Chemistry 2023; 16(1) , 147-158.

[29] A Purniawan, MI Lusida, RW Pujiyanto, AM Nastri, AA Permanasari, AAH Harsono, NH Oktavia, ST Wicaksono, JR Dewantari, RR Prasetya, K Rahardjo, M Nishimura, Y Mori and K Shimizu. Synthesis and assessment of copper-based nanoparticles as a surface coating agent for antiviral properties against SARS-CoV-2. Scientific Reports 2022; 12(1) , 4835.

[30] VNT La, S Nicholson, A Haneef, L Kang and DDL Minh. Inclusion of control data in fits to concentration-response curves improves estimates of half-maximal concentrations. Journal of Medical Chemistry 202; 66(18) , 12751-12761.

[31] L Grumiro, M Brandolini, G Gatti, A Scalcione, F Taddei, G Dirani, A Mancini, A Denicolò, M Manera, S Zannoli, MM Marino, M Morotti, V Arfilli, A Battisti, M Cricca and V Sambri. Evaluation of STANDARD ^TM M10 SARS-CoV-2, a novel cartridge-based real-time PCR assay for the rapid identification of severe acute respiratory syndrome coronavirus 2. Applied Microbiology 2022; 2(4) , 873-881.

[32] A Daina, O Michielin and V Zoete. SwissADME: A free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Scientific Reports 2017; 7 , 42717.

[33] S Abu-Melha, MM Edrees, MA Said, SM Riyadh, NS Al-Kaff and SM Gomha. Potential COVID-19 drug candidates based on diazinyl-thiazol-imine moieties: Synthesis and greener pastures biological study. Molecules 2022; 27(2) , 488.

[34] CA Lipinski, F Lombardo, BW Dominy and PJ Feeney. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Advanced Drug Delivery Reviews 2001; 46(1-3) , 3-25.

[35] R Yadav, M Imran, P Dhamija, DK Chaurasia and S Handu. Virtual screening, ADMET prediction and dynamics simulation of potential compounds targeting the main protease of SARS-CoV-2. Journal of Biomolecular Structure and Dynamics 2021; 39(17) , 6617-6632.

[36] L Vegosen and TM Martin. An automated framework for compiling and integrating chemical hazard data. Clean Technologies and Environmental Policy 2020; 22(2) , 441-458.

[37] TM Martin. User’s Guide for T.E.S.T. (Toxicity Estimation Software Tool) Version 5.1. 2020, Available at: https://www.epa.gov/chemical-research/toxicity-estimation-software-tool-test, accessed May 2025.

[38] JA Maier, C Martinez, K Kasavajhala, L Wickstrom, KE Hauser and C Simmerling. ff14SB: Improving the accuracy of protein side chain and backbone parameters from ff99SB. Journal of Chemical Theory and Computation 2015; 11(8) , 3696-3713.

[39] S Kumar and S Kumar. Molecular docking: A structure-based approach for drug repurposing . In : K Roy (Ed.). In silico drug design. Academic Press, Massachusetts, 2019, p. 161-189.

[40] MS Coumar. Molecular docking for computer-aided drug design: Fundamentals, Techniques, Resources and Applications . Academic Press, Massachusetts, 2021.

[41] WJ Allen, TE Balius, S Mukherjee, SR Brozell, DT Moustakas, PT Lang, DA Case, ID Kuntz and RC Rizzo. DOCK 6: Impact of new features and current docking performance. Journal of Computational Chemistry 2016; 36(15) , 1132-1156.

[42] S Khasimbi, F Ali, K Manda, A Sharma, G Chauhan and S Wakode. Dihydropyrimidinones scaffold as a promising nucleus for synthetic profile and various therapeutic targets: A review. Current Organic Synthesis 2021; 18(3) , 270-293.

[43] P Pérez-Ramos, G Biniari, RG Soengas, H Rodríguez-Solla and C Simal. Visible-light photocatalysis for sustainable chromene synthesis and functionalization. Chemistry 2025; 31(29) , e202500283.

[44] WW Mar, KU Haq, RA Rusdipoetra, RPFC Samiadji, A Rohman, NNT Puspaningsih and H Suwito. Anticancer activity through inhibition of BCL6 ^BTB of chalcone - Thiourea hybrid compounds: A molecular docking study. AIP Conference Proceedings 2023; 2679(1) , 50009.

[45] I Amalina, NNT Puspaningsih and H Suwito. In silico analysis of pyrimidine derivatives as potential antibacterial agents. AIP Conference Proceedings 2023; 2536 , 50004.

[46] G Indrayanto, GS Putra and F Suhud. Validation of in-vitro bioassay methods: Application in herbal drug research. Profiles of Drug Substances, Excipients and Related Methodology 2021; 46 , 273-307.

[47] AJ Pruijssers, AJ Pruijssers, AS George, A Schäfer, SR Leist, LE Gralinksi, KH Dinnon, BL Yount, ML Agostini, LJ Stevens, JD Chappell, X Lu, TM Hughes, K Gully, DR Martinez, AJ Brown, RL Graham, JK Perry, VD Pont, J Pitts, …, TP Sheahan. Remdesivir inhibits SARS-CoV-2 in human lung cells and chimeric SARS-CoV expressing the SARS-CoV-2 RNA polymerase in mice. Cell Reports 2020; 32(3) , 107940.

[48] M Wang, R Cao, L Zhang, X Yang, J Liu, M Xu, Z Shi, Z Hu, W Zhong and G Xiao . Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro . Cell Research 2020; 30(3) , 269-271.

[49] B Morak-Młodawska, M Jeleń, E Martula and R Korlacki. Study of lipophilicity and ADME properties of 1,9-diazaphenothiazines with anticancer action. International Journal of Molecular Sciences 2023; 24(8) , 6970.

[50] LM Castro-González, JR Alvarez-Idaboy and A Galano. Computationally designed sesamol derivatives proposed as potent antioxidants. ACS Omega 2020; 5(16) , 9566-9575.

[51] A Pérez-González, R Castañeda-Arriaga, EG Guzmán-López, LF Hernández-Ayala and A Galano. Chalcone derivatives with a high potential as multifunctional antioxidant neuroprotectors. ACS Omega 2022; 7(43) , 38254-38268.

[52] A Fischer, M Sellner, S Neranjan, M Smieško and MA Lill. Potential inhibitors for novel coronavirus protease identified by virtual screening of 606 million compounds. International Journal of Molecular Sciences 2020; 21(10) , 3626.

[53] Z Jin, X Du, Y Xu, Y Deng, M Liu, Y Zhao, B Zhang, X Li, L Zhang, C Peng, Y Duan, J Yu, L Wang, K Yang, F Liu, R Jiang, X Yang, T You, X Liu, …, H Yang. Structure of M ^pro from SARS-CoV-2 and discovery of its inhibitors. Nature 2020; 582 , 289-293.

[54] S Guo, H Xie, Y Lei, B Liu, L Zhang, Y Xu and Z Zuo. Discovery of novel inhibitors against main protease (Mpro) of SARS-CoV-2 via virtual screening and biochemical evaluation. Bioorganic Chemisty 2021; 110 , 104767.

[55] B Chatterjee and SS Thakur. Remdesivir and its combination with repurposed drugs as COVID-19 therapeutics. Frontiers in Immunology 2022; 13 , 830990.

[56] M Prajapat, Sarma, N Shekhar, P Avti, S Sinha, H Kaur, S Kumar, A Bhattacharyya, H Kumar, S Bansal and B Medhi. Drug targets for corona virus: A systematic review. Indian Journal of Pharmacology 2020; 52(1) , 56-65.

[57] A Shirali, V Stebliankin, U Karki, J Shi, P Chapagain and G Narasimhan. A comprehensive survey of scoring functions for protein docking models. BMC Bioinformatics 2025; 26 , 25.

[58] A Harmalkar and JJ Gray. Advances to tackle backbone flexibility in protein docking. Current Opinion in Structural Biology 2021; 67 , 178-186.

[59] GL Patrick. An introduction to medicinal chemistry . 7 ^th eds. Oxford University Press, Oxford, 2023.

[60] M Hariono, P Hariyono, R Dwiastuti, W Setyani, M Yusuf, N Salin and H Wahab. Potential SARS-CoV-2 3CLpro inhibitors from chromene, flavonoid and hydroxamic acid compound based on FRET assay, docking and pharmacophore studies. Results in Chemistry 2021; 3 , 100195.

[61] S Gorai, V Junghare, K Kundu, S Gharui, M Kumar, BS Patro, SK Nayak, S Hazra and S Mula. Synthesis of dihydrobenzofuro[3,2-b]chromenes as potential 3CLpro inhibitors of SARS-CoV-2: A molecular docking and molecular dynamics study. ChemMedChem 2022; 17(8) , e202100782.

[62] LH Al-Wahaibi, AM Elshamsy, TFS Ali, BGM Youssif, S Bräse, M Abdel-Aziz and NA El-Koussi . Design and synthesis of new dihydropyrimidine/sulphonamide hybrids as promising anti-inflammatory agents via dual mPGES-1/5-LOX inhibition. Frontiers in Chemistry 2024; 12 , 1387923.