An Improved Synthesis of Key Intermediate to the Formation of Selected Indolin-2-Ones Derivatives Incorporating Ultrasound and Deep Eutectic Solvent (DES) Blend of Techniques, for Some Biological Activities and Molecular Docking Studies.

We have developed a new idea to synthesize a key intermediate molecule by utilizing deep eutectic solvent (DES) and ultrasound in a multistep reaction to ensure process cost-effectiveness. To confirm the stability of reagents with DES, electronic energies were calculated at the B3LYP/6-31+G(d,p) level of theory. DES stabilized the reagents mainly due to strong intermolecular hydrogen bonding. Key intermediate (3) and final compounds (4a-n) were synthesized in a higher yield of 95% and 80%-88%, respectively. Further, final compounds (4a-n) were assessed for their anti-inflammatory, analgesic, ulcerogenic, and lipid peroxidation. The compounds 4f, 4g, 4j, 4l, and 4m showed good anti-inflammatory activity, while 4f, 4i, and 4n exhibited very good analgesic activity as compared to the standard drug. The ulcerogenicity of selected compounds was far less than the indomethacin. The ligands had also shown a good docking score (4f = -6.859 kcal/mol and 4n = -7.077 kcal/mol) as compared to control indomethacin (-6.109 kcal/mol) against the target protein COX-2. These derivatives have the potential to block this enzyme and can be used as NSAID. The state-of-art DFT theory was used to validate the lipid peroxidation mechanism of the active compounds which was in good agreement with the variations of BDEs and IP of the tested compounds.


Introduction
Nonsteroidal anti-inflammatory drugs (NSAIDs) are a profound application for the treatment of inflammatory diseases and pain. NSAIDs are the choice of treatment in various inflammation and pain-related problems such as osteoarthritis, rheumatoid arthritis, spondylitis, and gout [1][2][3]. A mechanism-based action of these drugs is exerted through the inhibition of cyclooxygenase type of enzymes, a principal enzyme that is used in the conversion of arachidonic acid to prostaglandin [4][5][6]. It has been reported that two forms of cyclooxygenase are involved in the pathogenesis of pain and inflammation, COX-1 and COX-2 [7,8]. However, their regulation and expression in the body are different [8,9], COX-1 is a known constitutive enzyme which helps in cytoprotection in the gastrointestinal tract (GI). The inhibition of COX-1 produces the undesired side effects of NSAIDs, for example, gastrointestinal toxicity because of their ulcerogenic effects. The COX-2 is an inducible enzyme that works through the mediation of the selective inflammatory signal and the therapeutic anti-inflammatory action of NSAIDs is produced by the inhibition of COX-2 [10][11][12][13][14]. Based on this observation, many selective COX-2 inhibitors such as celecoxib, rofecoxib, and valdecoxib emerged as relatively safe NSAID'S together with improved gastric problems. However, the reporting of the cardiovascular side effects, for example, increased risk of myocardial infarction, stroke, heart failure, and hypertension caused the withdrawal of many COX-2 inhibitors from the market [15]. This encouraged research professionals to develop newer chemical entities as anti-inflammatory agents with minimal side effects.
Indole ring and its derivatives have emerged as privileged pharmacophores representing more than thousands of natural isolates with known biological and pharmaceutical activities such as anti-inflammatory, analgesic [16][17][18][19], antimicrobial [20], antitumor [21], and anticonvulsant [22]. This ring is also a vital part of indomethacin, which is currently marketed as NSAIDs. However, the gastric safety profile of indomethacin is not promising and it produces gastrointestinal toxicity because of its ulcerogenic effects. In recent times, research reports highlighting the usefulness of the development of new coumarinylthiazoles as an anti-inflammatory and analgesic agents have also been published [23][24][25]. Thiazole and indole type of moieties were reported to synthesize by utilizing harsh chemicals/solvents which causes environmental pollution, as well as raise the risk of health issues [26,27]. An alternative to such solvents such as deep eutectic solvent (DES) is the most valuable choice for varieties of organic transformations [28,29]. DES is usually a mixture of compounds having melting points less than their mixing components. The most versatile DES was prepared from choline chloride and some hydrogen bond donor (urea, glycerol) [29]. Depression in the melting point of DES is associated with the molecular interaction of choline chloride and hydrogen bond donor part [29].
The immense application of ultrasound has been highlighted recently in organic and material science [30,31]. Ultrasound increased the rate of reaction by acoustic cavitation phenomena generated as a result of initiation, growth, and collapse of bubbles during the course of reactions.
Keeping these things and with extended work [32][33][34][35] of our group to the development of new chemical templates in order to discover novel NSAIDs, authors planned to synthesize some molecules with a low budget and utilizing deep eutectic solvent and ultrasound technique to fulfill green chemistry approach.
Other bands due to -C=N and two -C=O functional groups were found at around 1600 and 1700, respectively. In the 1 H-NMR spectra, two -NH peaks appeared at around 9 and 10 ppm. The lower value provides information as a singlet due to -NH attached as -CH 2 NH with indolinone nitrogen as a characteristic peak. Value at δ 5 ppm confirms the presence of -CH 2 which is another important peak for identification. Further, characteristics peak of -CH 2 of -CH 2 NH was confirmed by 13 C-NMR around δ 69 ppm.
The characterization data of all the synthesized compounds are provided below.
2-(2-Oxoindolin-3-ylidene)hydrazine carbothioamide (2)    To develop the efficient method as compared to conventional, we have conducted the synthesis of key intermediate (3) utilizing biocompatible deep eutectic solvent (DES) and ultrasound blend of technique. As a result of combined use of DES and ultrasound, an increase in % yield of key intermediate as high as 95% has been found with the expense of 1 h only. Whereas, a similar type of organic transformation using dioxane and another organic solvent together with conventional heating were reported to have % yield around 44%-68% in 3-4 h [36][37][38]. Further, we have also found 80%-88% of all final compounds (4a-n) utilizing ultrasound as a source of heating. Some of our earlier work and other related literature also mentioned the significance DES and ultrasound technology as an energy saving process [29,39,40] which is certainly a good favor of our present work.

Scheme 1. The proposed mechanism involved in the formation of key intermediate, 3-(2-(4-(2oxochroman-3-yl) thiazol-2-yl) hydrazono) indolin-2-one (3) using deep eutectic solvent (DES).
To emphasize the role of DES in the stability of starting reagents, the electronic energies of the latter were calculated at the B3LYP/6-31+G(d,p) level of theory in the presence and absence of DES ( Figure 1). The presence of DES stabilizes both reagents by 225 hartree (Figure 1). The stability of the starting material in the presence of DES is mainly referred to the formation of strong intermolecular hydrogen bonding between DES and the starting reagents ( Figure 1). Moreover, ultrasound also played a significant role in the formation of the desired compound. Under the influence of sonic waves inside the reaction vessel, there was the formation of microscopic bubbles, as a result of high temperature and pressure [28][29][30][31]. These tiny microscopic bubbles also help in the cyclization process.

Anti-Inflammatory Activity
The anti-inflammatory activity of the synthesized compounds (4a-n) was evaluated by the carrageenan-induced paw edema method. An oral dose of 10 mg/kg was used for compounds and compared with the standard. Anti-inflammatory activity was accessed through percentage inhibition after 2 and 4 h ( Table 1). Anti-inflammatory activity in terms of percentage inhibition for the test compounds is ranging from 5.57% to 80.94% (Table 1), whereas the standard drug showed 82.05% after 4 h. Compounds 4f (72.42%), 4g (77.94%), 4j (77.90%), 4k (70.75%), 4l (80.94%), and 4m (78.42%) showed comparable results against the standard drug.

Compound
Mean with ± SEM % Analgesic Activity (Dose = 10 mgkg −1 ) Potency All the tested compounds and standard drugs are evaluated at 10 mg/kg oral dose. It was identified that compound (4l) which showed maximum anti-inflammatory activity produces the least analgesic activity, but some selected compounds such as 4f, 4i, and 4n displayed analgesic activity in a similar fashion as the anti-inflammatory activity (Tables 1 and 2). Compound (4k) which exhibited the least analgesic activity was among the top-ranked anti-inflammatory activity. On the contrary, many compounds exhibiting good analgesic properties were not displayed as good anti-inflammatory activity and vice-versa (Tables 1 and 2).
After a close understanding of anti-inflammatory and analgesic potentials of compounds under the present series, we have made a structure-activity relationship. Compounds possessing a substituted phenyl ring showed better anti-inflammatory and analgesic activity than a compound having an unsubstituted phenyl ring. In most of the cases, the substitution of electron-withdrawing groups at C-2 and C-4 positions of the phenyl ring resulted in potent compounds except compound 4d (Ar = 4-bromophenyl) that showed negligible anti-inflammatory activity. Compound 4i possessing two electron-withdrawing groups exhibited moderate anti-inflammatory activity but good analgesic activity. Compound (4m) having an electron releasing group (-CH 3 ) at C-4 position exhibited better anti-inflammatory activity but less analgesic activity. On the other hand, a methyl group at C-2 showed better anti-inflammatory and analgesic agents (4n). A steep decrease in analgesic activity was observed when the phenyl ring was replaced by a triazole ring (4h).

Compound
Severity

Lipid Peroxidation
Gastrointestinal (GI) ulceration, bleeding, and renal problems are common complications of NSAID'S consumption, which is directly related to lipid peroxidation. It has been evidenced that drugs having less ulcerogenecity showed reduced malondialdehyde (a byproduct of lipid peroxidation) content [4,41]. We have examined the lipid peroxidation (LP) of compounds which exhibited maximum anti-inflammatory and analgesic activities (4c, 4f, 4i, 4n). It was measured as the nmol of MDA/100 mg of gastric tissue. We have found the Lipid peroxidation value was maximum 6.71 ± 0.18 for indomethacin, whereas 3.89 ± 0.17, 4.08 ± 0.22, 4.26 ± 0.12, and 4.81 ± 0.13 for compounds 4i, 4c, 4f, and 4n, respectively. It was interesting to mention that all these compounds having electron-withdrawing functionality on the phenyl ring (except 4n) exhibited less ulcerogenecity with reduced lipid peroxidation (Table 3).

DFT Results
As mentioned above, only the synthesized derivatives (In-H) that exhibited maximum anti-inflammatory and analgesic activities derivatives are subjected to the lipid peroxidation (LP) test (Table 4, Figure 2). The tested In-H derivatives show the ability to scavenge LOO• free radicals. To shed light on the small observed lipid peroxidation inhibition of In-H derivatives, bond dissociation enthalpies of the of i-NH function groups and ionization potential energies of the tested compounds were calculated at the B3P86/6-311+G(d,p) level of theory (Tables 3 and 4). The tested compounds showed similar lipid peroxidation with a small variation between their values. This result is confirmed by the small differences of BDEs of the active 17-NH group and IP energies, where the maximum variations of BDEs and IPs are 0.03 kcal/mol and 0.08 eV, respectively.

In Silico Study Target Protein Selection and Retrieval
The 3D structure of target protein COX-2 from two different organisms, i.e., mouse and human were retrieved from the protein data bank having PDB id 3NT1 and 5F91, respectively [42,43].

Protein (COX-2) Preparation and Validation
The protein structures obtained from PDB were modified suitably for the molecular docking studies. The modified protein structures were then validated through the Ramachandran plot. The Ramachandran plot of these two-target proteins is shown in Figure 3a,b.
The evaluation of phi/psi angles validates the prepared protein structures as most of the residues are in the most favored region. In the case of 3NT1, 90.8% amino acid residues are in the most favored regions whereas 8.8%, 0.1%, and 0.3% are in additional allowed regions, generously allowed regions, and disallowed regions, respectively. Similarly, in 5F91 90.7%, 9.1%, 0.1%, and 0.1% amino acid residues are in the most favored regions, additionally allowed regions, generously allowed regions, and disallowed regions, respectively. The result obtained allows the use of these structures for further molecular docking studies.

Prediction and Evaluation of the Binding Site in COX-2
The site map application predicts five different drug binding sites in both the target proteins. The site score for the protein 3NT1 was 1.078, 1.053, 1.048, 1.034, and 0.991. Similarly, the site score obtained for the protein 5F19 was 1.082, 1.051, 1.046, 1.034, and 0.990. As a rule of thumb, the binding sites having a score above one are considered as druggable pockets. In the present in silico study, the site with the highest score was selected for the molecular docking studies. The druggable pocket inside the respective target proteins is shown in Figure 4a

Ligand Preparation
The lowest energy conformation of each test ligand (4a-4n) was prepared for the docking studies as per the standard guidelines and used in the molecular docking studies.

Grid Generation in the Target Protein COX-2
After the determination of the exact location of the drug binding site in each target, the protein grid was generated around the binding sites to specify the volume and location of the druggable pocket.

Molecular Docking Studies
All the synthesized ligands (4a-n), i.e., the derivatives of 1-(substituted phenyl aminomethyl)-3-(2-(4-(2-oxochroman-3-yl) thiazol-2-yl) hydrazono) indolin-2-ones (4a-n) were docked inside the binding pocket of target proteins COX-2 (PDB ID: 3NT1 and 5F19) using Glide application (version 7.0, Schrödinger, New York, USA). To compare the docking efficacy of test ligands, indomethacin was used as a reference/control drug. The molecular docking results were evaluated on the basis of docking score, E-model score, and binding energy as summarized in Table 5. The maximum test ligands that are 4 a-c, e, g-n showed docking score lower than the control/reference drug (−6.324 kcal/mol) against the mouse target protein. A similar pattern of docking score is observed against human target protein where all test ligands (4a-n) have lower docking scores as compared to control (docking score −6.109 kcal/mol). It was believed that a low binding energy dock conformer exhibited maximum stability. The two best compounds on the basis of experimental results is 4n and 4f. 4n and 4f have docking score −7.077 and −6.859 kcal/mol against human target protein, respectively. The same two ligand 4n and 4f have a score −6.693 and −6.071 kcal/mol against mouse target protein, respectively. The docked ligands (4n and 4f) inside the binding pocket of the respective target proteins (3NT1 and 5F19) is shown in Figures 5 and 6.  The further efficacy of the docking is interpreted in terms of interaction that exists between the ligand and the surrounding amino acid residues inside the druggable pocket. The overall binding interaction (in terms of bonding) for each ligand is summarized in Tables 6 and 7 for the proteins 3NT1 and 5F19, respectively. Table 6. Type of interaction and amino acid residues involved in that interaction inside the binding pocket of the Cox-2 enzyme from mouse (4NT1).  Table 7. Type of interaction and amino acid residues involved in that interaction inside the binding pocket of the Cox-2 enzyme from mouse (5F19). Among the two potent ligands, 4n is more suitable for drug candidates as it possesses a strong affinity towards the target proteins. In 3NT1, it forms two hydrogen bonds with Arg 376, whereas in 5F19 two pi-pi stacking exists with the involvement of Phe 142 and Arg 333. In the case of 4f, there is no hydrogen bonding or pi-pi interaction is observed (whether it is 3NT1 or 5F19). All these interactions are shown in Figures 7 and 8 as a ligand interaction diagram. The molecular docking studies suggest that these two derivatives had potential to be used as a NSAID drug as it has the ability to block the COX-2 enzyme.

ADME Profiling
The suitability of test ligands as drug candidates depending on their pharmacokinetic behavior was also assessed using the in silico approach. Many drug candidates fail at a later stage of a clinical trial due to their poor pharmacokinetic performance. In order to avoid such failure and save time, energy, and money, in silico ADME profiling is a good choice [44]. The result of in silico ADME profiling is presented in Table 8. The result obtained suggests that the values of test parameters are within the recommended range and can be used as a drug candidates (https://www.mdpi.com/2504-3900/41/1/8).
The oral drug absorption is predicted in terms of apparent Caco-2 permeability (QPPlogCaco) that represents the gut-blood barrier. The value above 500 indicates a great absorption while below 25 is considered a poor score [44]. The ligand 4n and 4f have QPPlogCaco value 619.284 and 479.473 that is very good as compared to indomethacin that has a score of 185.783 only. The Madin-Darby canine kidney (MDCK) cell model is used to investigate the apparent MDCK cell permeability [45]. The score above 500 is considerably good that is obtained in the case of both the potent test ligands. The apparent MDCK permeability score for 4n and 4f is 586.303 and 809.359 whereas the standard drug has a value of 251.855. The percent human abortion of both the potential ligand is also comparable to the standard and above 80%. The test ligands are also found to be obeying the Lipinski rule of 5. Table 8. Absorption distribution metabolism and excreation (ADME) profiling of all the ligands (4a-n) synthesized to be used as a drug candidate.

Statistical Analysis
Data used in the experimental pharmacological section was used as the mean ± standard error (SEM). One way analysis of variance (ANOVA) and Dennett's multiple comparison test techniques was employed to compare between the test, control, and standard group, utilizing statistical software GraphPad Prism version 5.00, California corporation, San Diego, USA. Such results showed significantly different at p < 0.05.

Materials and Methods
Melting points were evaluated in open capillary tubes and are uncorrected. 5 PC FTIR spectrometer (Thermo scientific, New York, NY, USA, Bruker DRX-300 FT NMR, Bruker, Karlsruhe, Germany) spectrophotometer and Jeol-JMS-D-300 mass spectrometer (70 eV) (Jeol, Tokyo, Japan) for IR, NMR, and mass, respectively were used to characterize the compounds.

Preparation of Deep Eutectic Solvent (DES)
A mole ratio (1:2) of choline chloride and urea were chosen to prepare DES as per the reported method [28]. (3) In a specially designed sonicating flask, an equimolar (0.01 mole) quantity of 3-bromoacetyl coumarin and 2-(2-oxoindolin-3-ylidene) hydrazinecarbothioamide (2) with 8 g of prepared DES was added. A sonicating probe of 26 kHz frequency at 40% amplitude was submerged into the reaction vessel. The completion of the reaction was monitored by taking TLC at regular intervals. Upon completion, it was poured onto crushed ice. Then, upon completion of the reaction, it was extracted by dichloromethane using a separating funnel. The organic solvent layer was collected and evaporated to get the desired product. DES was isolated and kept for future use.

Preparation of 2-(2-Oxoindolin-3-Ylidene) Hydrazine Carbothioamide (2)
Compounds produced were assessed for their anti-inflammatory activity using the carrageenaninduced hind paw edema method [45]. The anti-inflammatory activity was carried out using Wistar albino rats of either sex (150-220 g) using a Digital Plethysmometer (Model No. 7140, UGO BASILE). The edema was induced by using 1% carrageenan solution. Indomethacin was used as a standard drug. The anti-inflammatory activity of the standard drug and tested compounds was determined at a dose of 10 mg/kg body weight. The animals were divided into groups containing six animals each and the initial paw volume of each rat was noted by the NaCl displacement method. One group was kept as control, one as standard, and the rest of the groups of compounds were to be tested. To the control group, 1% CMC solution was administered p.o. To the standard group, the standard drug was administered orally. To the test group, tested compounds were administered orally. After 60 min of the 1% CMC solution/standard drug/test compound administration, 0.1 ml of 1% (w/v) carrageenan was injected in the plantar region of the hind limb (right) of all the rats in each group including the control group. The paw volume was again measured after the time interval of 2 and 4 h. Using the following formula, inflammation was calculated as percentage inhibition for the test and reference compounds (Final foot volume of control-final foot volume of std./test) × 100 / final foot volume of control)

Analgesic Activity
The analgesic activity of the tested compounds was carried out by acetic acid-induced writhing method as given in the literature [9] using Swiss albino mice of either sex (25-35 g). The writh were induced in the albino mice using an intraperitoneal injection of 1% acetic acid solution. The standard drug indomethacin and test compounds were evaluated at a concentration of 10 mg/kg of the body weight. The animals were divided into groups and each group consisted of six animals. One group was kept as control, one as standard, and other as test groups. To the control group, 0.1% CMC solution was administered p.o; to the standard group, the standard drug was administered orally, and to the test group, test compounds were administered orally. After 60 min of the 0.1% CMC solution/standard drug/tested compound administration, 0.1 ml of 1% (v/v) acetic acid solution in distilled water was injected intraperitoneally in all the mice in each group including the control group. The writhing (a contraction of the abdomen, turning of trunk, and extension of hind limbs) was counted after 5 min of acetic acid administration and were counted for a period of 15 min. The percentage of analgesic activity was calculated using the following formula.

Acute Ulcerogenic Activity
Acute ulcerogenic activity evaluation of the synthesized compounds was carried out according to the method described [46] using Wistar rats of either sex (180-220 g). The animals were distributed into control, group, and test group. Each group consisted of six rats. All the rats fasted for 24 h with free access to water. To the control group, 1% CMC solution was administered p.o; to the standard group, indomethacin at a concentration of 20 mg/kg was administered orally; and to the test groups, tested compounds were administered orally at a concentration of 30 mg/kg. After the dose administration animals were kept for 17 h. After this, the animals were sacrificed for the appraisal of ulcerogenic assessment. The stomach was taken out from the animal body and washed with flushing water, then with a cotton swab wetted with saline (0.9%) and pinned on wax-coated try. The glandular portion of the stomach was cleaned again with saline to closely identify the presence of a type of ulcers or hemorrhage mark using a magnifying glass. The mucosal injury of the stomach was evaluated as per the following system: 0.5 = redness; 1 = spot ulcer; 1.5 = hemorrhage streak; 2 = ulcers < 3; 3 = ulcers > 3 < 5. The value obtained as a result of the mean score of the individual treated group-mean score of control is referred to as the severity index of the gastric mucosal damage.

Lipid Peroxidation Study
The method adopted for lipid peroxidation is the same as Ohkawaet al. [47] and the recent work of our researchers [9].

Theoretical Details
It is well known that almost all phenolic compounds may inhibit the lipid peroxidation process due to their ability to scavenge the chain-carrying lipid peroxyl radicals, LOO•. The lipid, LH, peroxidation process is represented by three main steps initiation, propagation, and terminations. The scavenging of LOO• by the synthesized indolin-2-Ones derivatives (In-H) may refer to hydrogen atoms transferred or an electron transfer from the former to the lipid peroxyl radical. The hydrogen atom transfer is represented by the following reaction: The above lipid peroxidation inhibition is governed by bond dissociation enthalpies (BDE) of i-NH groups of the synthesized indolin-2-ones derivatives (In-H). BDE is calculated using the following equation: The minimum value of BDE indicates that hemolytic bond dissociation is much easier, which is helpful in the lipid peroxidation process.
Previously, we showed the success of the hybrid functional B3B86 in rationalizing the scavenging of free radicals by synthesized natural polyphenols [48][49][50]. Hence, we extended here the use of B3P86 to the In-H synthesized derivatives as lipid peroxyl radical inhibitors. We have already tested, the basic set effect on BDEs of hispidin and isohispidin isomers by using varieties of basic sets. The obtained BDEs showed differences lower than 0.4 kcal/mol for active sites and a slight influence on IP values [48]. Consequently, a double basis set, 6-31+G(d,p), was used in this study. The 3D geometry optimization of In-H derivatives and their corresponding radicals In• were performed at the B3P86/6-6-31+G(d,p) level of theory. The ground state minima were confirmed by vibrational frequency calculations (i.e., the absence of imaginary frequencies). All DFT chemical calculations have been performed using the mentioned methodology, as implemented in the Gaussian 09 package [51].

Software
All the in silico experiments of the present study were carried out using the Schrodinger Maestro interface (Maestro, version 10.5, Schrödinger, LLC, New York, NY, USA, 2016) [52,53].

Target Protein Selection and Retrieval
In the present study, COX-2 is selected as the target protein because the therapeutic response of NSAID's is generated by blocking/inhibiting this enzyme. The 3D structure of COX-2 was retrieved from Protein Data Bank (PDB, https://www.rcsb.org/). There are two different structures of this enzyme available, one from the mouse and another from the human. The two structures obtained have PDB ID 3NT1 (mouse) and 5F91 (human).

Protein (COX-2) Preparation and Validation of Prepared Structure
The 3D structure of target proteins obtained from PDB was prepared (for further steps) using the protein preparation wizard (version 4.3, Schrödinger, New York, NY, USA). The protein preparation is a multi-step process that includes the addition of hydrogen atoms, optimization of hydrogen bonds, and elimination of any atomic level clashes. The final step of protein preparation is energy minimization that was performed at the condition 0.3 Å of RMSD and OPLS_2005 force field [54]. The protein structures prepared were further validated through the Ramachandran plot that is based on phi/psi angles evaluation.
Prediction and Evaluation of the Binding Site in COX-2 To locate the ligand-binding site in the target protein, site map application (version 3.8, Schrödinger, New York, NY, USA) was used. The potency of the predicted site is decided on the basis of the site score generated by the tool.
The binding site effectiveness is determined by several physical parameters such as size, the degree of enclosure/exposure, hydrophobic/hydrophilic character, opportunities of hydrogen bonding, etc.
Grid Generation in the Target Protein COX-2 The grid was created nearby the binding site in the respective target proteins. It determines the exact position and size of the binding site in terms of receptor grids that are required for the molecular docking step. The box size taken is of dimension 20 × 20 × 20 Å and the atoms were scaled by van der Waals radii of 1.0 Å having partial atomic charge less than 0.25.

Docking of Ligands and COX-2
The prepared ligands were docked with the COX-2 (target protein) at the respective binding site using the Glide (version 7.0 Schrödinger, New York, NY, USA) tool. The extra precision (XP) algorithm was employed for the docking operation and output is obtained in the form of docking score. It determines a possible binding pose between the target and the ligand. It also provides information about the most favorable interactions among them [55][56][57].

Conclusions
In conclusion, an improved synthesis of key intermediate through the combined use of deep eutectic solvent and ultrasound is a rational approach to enhance the yield of desired compounds via an economically viable and environmentally acceptable way. The stability of reagents with DES was verified by calculating electronic energy. Further, all the final compounds (4a-n) have been evaluated as anti-inflammatory and analgesic activities. Selected compounds were further tested for ulcerogenic and lipid peroxidation potential. Only two compounds claimed to be most potent as anti-inflammatory and analgesic molecule with the highest reduction in GI toxicity. In silico study also supports the utility of these two potent ligands as drug candidates and paves the path for future drug development studies. The active compounds showed similar lipid peroxidation activities, and this was mainly due to their closest BDEs and IP values, i.e., the active compounds have the same potency to inhibit lipid radical by a hydrogen atom transfer from the active site of titled compounds to a lipid radical.