Design, Synthesis and In Vitro Evaluation of Spirooxindole-Based Phenylsulfonyl Moiety as a Candidate Anti-SAR-CoV-2 and MERS-CoV-2 with the Implementation of Combination Studies

The search for an effective anti-viral to inhibit COVID-19 is a challenge for the specialized scientific research community. This work investigated the anti-coronavirus activity for spirooxindole-based phenylsulfone cycloadducts in a single and combination protocols. The newly designed anti-SARS-CoV-2 therapeutics spirooxindoles synthesized by [3 + 2] cycloaddition reactions represent an efficient approach. One-pot multicomponent reactions between phenyl vinyl sulfone, substituted isatins, and amines afforded highly stereoselective anti-SARS-CoV-2 therapeutics spirooxindoles with three stereogenic centers. Herein, the newly synthesized spirooxindoles were assessed individually against the highly pathogenic human coronaviruses and proved to be highly potent and safer. Interestingly, the synergistic effect by combining the potent, tested spirooxindoles resulted in an improved antiviral activity as well as better host-cell safety. Compounds 4i and 4d represented the most potent activity against MERS-CoV with IC50 values of 11 and 23 µM, respectively. Both compounds 4c and 4e showed equipotent activity with the best IC50 against SARS-CoV-2 with values of 17 and 18 µM, respectively, then compounds 4d and 4k with IC50 values of 24 and 27 µM, respectively. Then, our attention oriented to perform a combination protocol as anti-SARS-CoV-2 for the best compounds with a different binding mode and accompanied with different pharmacophores. Combination of compound 4k with 4c and combination of compounds 4k with 4i proved to be more active and safer. Compounds 4k with 4i displayed IC50 = 3.275 µM and half maximal cytotoxic-concentration CC50 = 11832 µM. MD simulation of the most potential compounds as well as in silico ADMET properties were investigated. This study highlights the potential drug-like properties of spirooxindoles as a cocktail anti-coronavirus protocol.


Introduction
Since 2012, the world has been confronted with two highly pathogenic coronaviruses including the Middle East respiratory syndrome coronavirus (MERS-CoV) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1,2]. MERS is a viral respiratory infection that is caused by MERS-CoV, leading to severe disease with high mortality rates (approximately 35%) [1]. More recently, in 2019, a novel coronavirus (CoV) outbreak was

Synthesis of the Spirooxindole-Based Phenylsulfones 4a-n
The design, construction, and synthesis of new materials with significant antiviral applications towards COVID-19 are a challenge. In this text we employed the one potmulti component [3+2] cycloaddition (32CA) reaction approach for the synthesis of spirooxindole-based phenylsulfones as new materials [44]. The synthetic route is shown in Scheme 1. The spirooxindole-based phenylsulfone cycloadducts were obtained via 32CA reaction of phenyl vinyl sulfone 1 as the ethylene with the generated azomethine ylides (AYs) by reaction of many substituted isatins 2a-h (Isatin 2a, 5-Chloroisatin 2b, 5- To explore the phenylsulfone moiety in the spirooxindole framework and then to examine their activity against coronavirus is the source of a lot of attention for us. The goal of a combination protocol in a synergetic action is to combine medications that function through different or the same mechanisms, reducing the chances of side effects and developing resistance.
On the other hand, it is clinically recommended to combine two or more drugs as a cocktail protocol approach for the treatment of viral infections [34][35][36][37]. In addition, the use of an effective combinational protocol could reduce the effective concentration of compounds below the therapeutic plasma concentrations, providing better clinical benefits.
S. Yuan and H. Sun, as co-workers, demonstrated one representative example for an orally administered cocktail therapy for SARS-CoV-2 viral infection treatment. This preclinical cocktail therapy consisted of N-acetyl-L-cysteine combined with colloidal bismuth subsalicylate (BSS) or bismuth sub-citrate (CBS) assessed in vivo, and exhibited high efficacy against a wide range of coronavirus replications such as the recently circulating SARS-CoV-2 Alpha variant (B.1.1.7), the low pathogenic human coronavirus 229E (HCoV-229E), and the Middle East respiratory syndrome coronavirus (MERS-CoV), blocking or inhibiting several cell-based and viral-based targets including angiotensin-converting enzyme 2 (ACE2), helicase (Hel), main protease (Mpro), and papain-like protease (PLpro) [38].
Based on these findings and continuing with our research program for drug discoveries [39][40][41][42][43][44][45][46], here we report on the design, synthesis, and anti-viral evaluation of a new spirooxindole with a phenylsulfonyl moiety as a promising lead compound with high efficacy and a safer cocktail protocol to block and inhibit the outbreaks of a new coronavirus disease.

Biological Studies
To test the preliminary antiviral activity of the synthesized spiro compounds against the highly pathogenic coronaviruses, SARS-CoV-2 "NRC-03-nhCoV" and MERS-CoV "NRCE-HKU270", the half maximal cytotoxic "CC 50 " and virus-inhibitory (IC 50 ) concentrations were determined using MTT assay and plaque reduction assay, respectively (Table 3). Except for the compounds 4b, 4f, 4g, 4j, and 4i-n, the tested spiro-compounds showed high to moderate antiviral activity against SARS-CoV-2 ranging from 17 to 37 µM against NRC-03-nhCoV and 15 to 74 µM against NRCE-HKU270 (Table 3). Interestingly, 4c, 4i, and 4k showed the best selectivity index (SI > 10) against SARS-CoV-2 and MERS-CoV in VERO E6 cells (Table 3). These results compared with remdesivir as a drug control that the FDA approved as anti-SARS-CoV-2, but it had more cytotoxic effects and side effects on patients so there remains an urgent need to present alternatives against anti-SARS-CoV-2 with higher safety and lower side effects. azomethine ylide (AY), then reacted with the sulfone derivatives to afford the final product. The chemical architecture was assigned based on a number of spectrophotometric tools including 1 HNMR, and 13 CNMR spectral analysis. The stereochemical and regio-specific outcomes of the 32CA reactions were confirmed by HNMR and X-ray single diffraction analysis for the compound 4m which has been published in our reported article [47,48]. Scheme 1. Synthesis of the spirooxindole-based phenylsulfone 4a-n.

Combination Protocol
A drug combination protocol using anti-viral drugs provides several advantages as it will be more effective than monotherapy due to the synergistic effect of complementary drugs, have lower side effects, and lower toxicity due to reducing the doses. For these reasons, the use of combinations of drugs which may ultimately have a positive impact on alleviating COVID-19 severity have been studied [49,50]. Most drug combinations against COVID-19 have included the in vitro investigation of the recommended doses of FDAapproved drugs in 1:1 or 1:0.5 combinations [49,51]. For newly synthesized compounds, rare studies have considered investigating the synergistic or antagonistic effect of the compounds in combination against COVID-19.
To further validate the anti-SARS-CoV-2 activity of the compounds 4c, 4i, and 4k, colorimetric crystal violet assay was used as previously described [52]. After equal concentrations from each compound were prepared (10 mg/mL), the mixture was prepared with an equal volume from each compound (1:1). Accordingly, the compound 4k showed the best IC 50 values. Therefore, we included it in combination with the other two safe/active compounds (4c and 4i). Interestingly, combinations with 4k improved the IC 50 of the compounds 4c and 4i. In addition, the mixture's CC 50 values improved when 4k was used in combination with 4c, and 4i compared to each individual compound. Based on these results, the selectivity indices for the combinations 4c/4k and 4i/4k were remarkably improved, to be 698.4 and 3612.8, respectively ( Figure 2). it will be more effective than monotherapy due to the synergistic effect of complementary drugs, have lower side effects, and lower toxicity due to reducing the doses. For these reasons, the use of combinations of drugs which may ultimately have a positive impact on alleviating COVID-19 severity have been studied [49,50]. Most drug combinations against COVID-19 have included the in vitro investigation of the recommended doses of FDA-approved drugs in 1:1 or 1:0.5 combinations [49,51]. For newly synthesized compounds, rare studies have considered investigating the synergistic or antagonistic effect of the compounds in combination against COVID-19.
To further validate the anti-SARS-CoV-2 activity of the compounds 4c, 4i, and 4k, colorimetric crystal violet assay was used as previously described [52]. After equal concentrations from each compound were prepared (10 mg/mL), the mixture was prepared with an equal volume from each compound (1:1). Accordingly, the compound 4k showed the best IC50 values. Therefore, we included it in combination with the other two safe/active compounds (4c and 4i). Interestingly, combinations with 4k improved the IC50 of the compounds 4c and 4i. In addition, the mixture's CC50 values improved when 4k was used in combination with 4c, and 4i compared to each individual compound. Based on these results, the selectivity indices for the combinations 4c/4k and 4i/4k were remarkably improved, to be 698.4 and 3612.8, respectively ( Figure 2).
Taking into consideration that remdesivir is one of the most promising anti-COVID-19 drugs, as demonstrated by previous in vitro and clinical studies [2,53,54], the inhibitory concentrations for the selected combinations 4c/4k and 4i/4k were lower than that for the control remdesivir drug (IC50 = 6.721 µ M). Interestingly, the IC50 values for the selected combinations were also potent compared to the other FDA-approved drugs that are applied in COVID-19 treatment protocols including viral protease inhibitors Lopinavir    50 ) in Vero-E6 cells and half maximal inhibitory concentration (IC 50 ) of the tested compounds 4c, 4i, and 4k, individually and in combination, against NRC-03-nhCoV in Vero-E6 cells, compared with remdesivir as a drug control. The 50% inhibitory concentration (IC 50 ) of each tested compound was calculated using nonlinear regression analysis in triplicate for each concentration used. The best fitting line was drawn between log concentrations and viral inhibition % using Graph Pad Prism software.
Taking into consideration that remdesivir is one of the most promising anti-COVID-19 drugs, as demonstrated by previous in vitro and clinical studies [2,53,54], the inhibitory concentrations for the selected combinations 4c/4k and 4i/4k were lower than that for the control remdesivir drug (IC 50 = 6.721 µM). Interestingly, the IC 50 values for the selected combinations were also potent compared to the other FDA-approved drugs that are applied in COVID-19 treatment protocols including viral protease inhibitors Lopinavir The compounds exhibited different binding modes and poses against the target protein.
According to the binding mode and site of interaction with the receptor, all compounds are classified in to three categories:

1.
Compounds 4n, 4b, 4e, 4m, 4f, 4j, and 4i have the same binding mode and these compounds were originated from secondary amino acids L-proline 3a, and L-thioproline 3b and were docked with complete overlay and with the detection of hydrogen bond (HB) with Arg: 116A through the carbonyl of oxoindoline moiety, Figure 3a (left domain of receptor); 2.
These compounds connected with same amino acids cleft with complete overlay and similarity without detection of any hydrogen bonds (HBs), Figure 3a (right domain of receptor); 3.
Compounds 4d, 4l, 4h, and 4c exhibited the same binding mode and pose with formation of HB with Arg: 33A through the hydroxyl functionality of pyrrolidine ring, Figure 3b. These compounds were synthesized from the secondary amino acids L-proline 3a and 2R,4R)-4-hydroxypyrrolidine-2-carboxylic acid (3c). Compound 4h formed HB with Thr:120A through NH of oxoindoline moiety, Figure S28. Compound 4l illustrated specific binding pose through hydrophobic-hydrophobic interaction, Figure S29. The compounds exhibited different binding modes and poses against the target protein. According to the binding mode and site of interaction with the receptor, all compounds are classified in to three categories: 1. Compounds 4n, 4b, 4e, 4m, 4f, 4j, and 4i have the same binding mode and these compounds were originated from secondary amino acids L-proline 3a, and Lthioproline 3b and were docked with complete overlay and with the detection of hydrogen bond (HB) with Arg: 116A through the carbonyl of oxoindoline moiety, Figure 3a (left domain of receptor); 2. Compounds 4a, 4g, and 4k were prepared from secondary amino acid L-proline 3a.
These compounds connected with same amino acids cleft with complete overlay and similarity without detection of any hydrogen bonds (HBs), Figure 3a (right domain of receptor); 3. Compounds 4d, 4l, 4h, and 4c exhibited the same binding mode and pose with formation of HB with Arg: 33A through the hydroxyl functionality of pyrrolidine ring, Figure 3b. These compounds were synthesized from the secondary amino acids L-proline 3a and 2R,4R)-4-hydroxypyrrolidine-2-carboxylic acid (3c). Compound 4h formed HB with Thr:120A through NH of oxoindoline moiety, Figure S28. Compound 4l illustrated specific binding pose through hydrophobic-hydrophobic interaction, Figure S29. The combination study was designed between compound 4k (second category) with compound 4i (first category); and compound 4k (second category) with compound 4c (third category). Interestingly, combinations with 4k improved the IC50 of the compounds 4c and 4i. In addition, the mixture's CC50 values were improved. From Figure 4a, compound 4k (grey color) binds with the amino acid cleft which differs from those interacting with compound 4i (green color). Compound 4i located in the site of the receptor close to the binding region of co-crystalized ligand. They formed HB with Arg: 116A and His:99A, respectively. Concerning the combination (compound 4k and 4c), compound 4c (green color) adopted the same region as shown before from compound 4i The combination study was designed between compound 4k (second category) with compound 4i (first category); and compound 4k (second category) with compound 4c (third category). Interestingly, combinations with 4k improved the IC 50 of the compounds 4c and 4i. In addition, the mixture's CC 50 values were improved. From Figure 4a, compound 4k (grey color) binds with the amino acid cleft which differs from those interacting with compound 4i (green color). Compound 4i located in the site of the receptor close to the binding region of co-crystalized ligand. They formed HB with Arg: 116A and His:99A, respectively. Concerning the combination (compound 4k and 4c), compound 4c (green color) adopted the same region as shown before from compound 4i but in a different binding mode and pose, Figure 4b. Compound 4c formed weak HB with Arg: 33A through its hydroxyl functionality of the pyrrolidine ring, Figure 4b   In order to examine the activity of these compounds against the MERS-CoV virus, the docking protocol was employed here against the main protease of MERS-CoV (PDB ID: 4ylu [56]). Both compound 4h and 4i were the most potent derivatives. Compound 4i docked with the receptor with hydrophobic-hydrophobic interaction in the same amino acids' clefts interacted with the co-crystalized ligand, Figure 5a, however compound 4h participated in a docking pose in a different domain, Figure 5b.

ADMET Analysis
Due to adverse pharmacokinetic properties, many of the prospective drug candidates never reach clinical trials. To examine the fundamental pharmacokinetic features of a compound, in silico ADMET analysis provides a valid alternative to earlier stage

Docking Study with MERS-CoV Viral Proteins nsp5 (PDB:ID: 4ylu)
In order to examine the activity of these compounds against the MERS-CoV virus, the docking protocol was employed here against the main protease of MERS-CoV (PDB ID: 4ylu [56]). Both compound 4h and 4i were the most potent derivatives. Compound 4i docked with the receptor with hydrophobic-hydrophobic interaction in the same amino acids' clefts interacted with the co-crystalized ligand, Figure 5a, however compound 4h participated in a docking pose in a different domain, Figure 5b.

ADMET Analysis
Due to adverse pharmacokinetic properties, many of the prospective drug candidates never reach clinical trials. To examine the fundamental pharmacokinetic features of a compound, in silico ADMET analysis provides a valid alternative to earlier stage

ADMET Analysis
Due to adverse pharmacokinetic properties, many of the prospective drug candidates never reach clinical trials. To examine the fundamental pharmacokinetic features of a compound, in silico ADMET analysis provides a valid alternative to earlier stage experiments to increase the success rate of clinical development. The bioavailability, pharmacokinetics, and toxicity of 4c, 4i, and 4k were attained by using ADMETlab and the obtained results are summarized in Table 4. The bioavailability and physiochemical properties were evaluated by plotting a radar representing 13 properties ( Figure 6). It is interesting to note that all of the properties were within their optimal ranges, which showed that 4c, 4i, and 4k had good oral bioavailability and druggability. Similarly, all of the compounds were predicted to follow the Lipinski rule of five with high gastrointestinal absorption. To further evaluate the toxicity and cross-reactivities of 4c, 4i, and 4k, the acute toxicity and PAINS were predicted. All of the three compounds were found to be non-toxic and non-cross-reactive. Taken together, 4c, 4i, and 4k were in significant agreement with the given criteria to be considered as drug-like. experiments to increase the success rate of clinical development. The bioavailability, pharmacokinetics, and toxicity of 4c, 4i, and 4k were attained by using ADMETlab and the obtained results are summarized in Table 2. The bioavailability and physiochemical properties were evaluated by plotting a radar representing 13 properties ( Figure 6). It is interesting to note that all of the properties were within their optimal ranges, which showed that 4c, 4i, and 4k had good oral bioavailability and druggability. Similarly, all of the compounds were predicted to follow the Lipinski rule of five with high gastrointestinal absorption. To further evaluate the toxicity and cross-reactivities of 4c, 4i, and 4k, the acute toxicity and PAINS were predicted. All of the three compounds were found to be non-toxic and non-cross-reactive. Taken together, 4c, 4i, and 4k were in significant agreement with the given criteria to be considered as drug-like.

Molecular Dynamics Simulation
The compounds 4c and 4k with the best selectivity index and significant antiviral activity against SARS-CoV-2 were subjected to molecular dynamics (MD) simulation to evaluate the time-dependent dynamics and stability of protein-ligand complexes. The docked pose of 4c and 4k in complex with RdRp of SARS-CoV-2 were subjected to 100ns of simulation and the Root Mean Square Deviation (RMSD) and Root Mean Square Fluctuation (RMSF) were calculated (Figure 7). The RMSD plot of 4k-RdRp showed variable fluctuations throughout the 100ns of simulation with the average RMSD of 4.3 ± 0.70 nm. The 4c-RdRp projected the more stable RMSD in comparison with the 4k complex with an average RMSD of 3.8 ± 0.47 nm. The 4c-RdRp complex was converged after 55ns and remained stable till the end of the simulation. To further evaluate the ligand-induced flexibility of SARS-CoV-2 RdRp residues, RMSF was calculated. Consistent with the RMSD results, the RSMF plot of the 4c complex showed a lesser magnitude of fluctuations as compared to the 4k complex. The average RMSF for the 4k-RdRp and 4c-RdRp complexes was found to be 6.04 ± 2.23 nm and 5.05 ± 1.78 nm, respectively. By analyzing the results, it was inferred that the experimental and theoretical results were consistent with one another.

Molecular Dynamics Simulation
The compounds 4c and 4k with the best selectivity index and significant antiviral activity against SARS-CoV-2 were subjected to molecular dynamics (MD) simulation to evaluate the time-dependent dynamics and stability of protein-ligand complexes. The docked pose of 4c and 4k in complex with RdRp of SARS-CoV-2 were subjected to 100ns of simulation and the Root Mean Square Deviation (RMSD) and Root Mean Square Fluctuation (RMSF) were calculated (Figure 7). The RMSD plot of 4k-RdRp showed variable fluctuations throughout the 100ns of simulation with the average RMSD of 4.3 ± 0.70 nm. The 4c-RdRp projected the more stable RMSD in comparison with the 4k complex with an average RMSD of 3.8 ± 0.47 nm. The 4c-RdRp complex was converged after 55ns and remained stable till the end of the simulation. To further evaluate the ligand-induced flexibility of SARS-CoV-2 RdRp residues, RMSF was calculated. Consistent with the RMSD results, the RSMF plot of the 4c complex showed a lesser magnitude of fluctuations as compared to the 4k complex. The average RMSF for the 4k-RdRp and 4c-RdRp complexes was found to be 6.04 ± 2.23 nm and 5.05 ± 1.78 nm, respectively. By analyzing the results, it was inferred that the experimental and theoretical results were consistent with one another.

General
The 1 H-NMR and 13 C-NMR spectra of 4a-n were recorded on a JEOL 400-MHz spectrometer (JEOL, Ltd, Tokyo, Japan) at ambient temperature. DMSO-d6 was used as the solvent; the chemical shifts (δ) are given in ppm.

Synthesis of the Spirooxindole-Based Phenylsulfone 4a-n (GP1)
A mixture of phenyl vinyl sulfone 1 (0.5 mmol), isatin derivatives 2a-h (0.5 mmol), and amino acids 3a-c (0.5 mmol) in methanol (10 mL) was refluxed in an oil bath for the appropriate time of 8 h. After completion of the reaction, as evident from TLC, the reaction was maintained at room temperature overnight, and the solid crystalline product was filtered off without any further purification.

General
The 1 H-NMR and 13 C-NMR spectra of 4a-n were recorded on a JEOL 400-MHz spectrometer (JEOL, Ltd, Tokyo, Japan) at ambient temperature. DMSO-d 6 was used as the solvent; the chemical shifts (δ) are given in ppm.

Biological Activity Assays
The protocol for the biological activity assay is provided in the Supplementary Materials.

Molecular Docking
The protocol for the molecular docking study is provided in the Supplementary Materials.