Lead optimization study on indoline-2,3-dione derivatives as potential fatty acid amide hydrolase inhibitors

Abstract Based on the known isatin-based fatty acid amide hydrolase (FAAH) inhibitor BSS-7, we designed and synthesized two small sets (6–13 and 17–20) of N-1 and C-3 substituted isatin derivatives and evaluated them for their in vitro FAAH inhibition properties. The lead simplification by modification of bulky aryl moiety at N-1 with a flexible allyl group produced a nanomolar (IC50 = 6.7 nM, K i = 5 nM) inhibitor 11 (Z)-3-((1H-benzo[d]imidazol-2-yl)imino)-1-allylindolin-2-one which exhibited a reversible and competitive FAAH inhibition with 1500 times more potency to BSS-7 (1.49 ± 0.03 µM). The lead compound 11 also showed a high blood–brain permeability and a significant antioxidant profile with no neurotoxicity. Docking results suggested that the inhibitor molecules occupied the active site of FAAH and offered optimal binding interactions. A molecular dynamics simulation study ascertained the stability of the lead inhibitor 11-FAAH complex. In silico ADMET profiling studies unveiled that compound 11 possesses good drug-like properties and merits further evaluation. Communicated by Ramaswamy H. Sarma


Introduction
Fatty acid amide hydrolase (FAAH, EC 3.5.1.4) is an endocannabinoid degrading enzyme that converts the endogenous substrate, anandamide into arachidonic acid and ethanolamine.It is a member of the amidase signature family and is characterized by the catalytic triad Ser217-Ser241-Lys142 in the active site (Giang & Cravatt, 1997;McKinney & Cravatt, 2003).FAAH is a primary enzyme responsible for the catabolism of fatty acid amides, and inhibition of FAAH is a viable approach to treat various CNS disorders like depression, anxiety, neuropathic pain, etc. (Jaiswal et al., 2022).The FAAH inhibition increases the endogenous anandamide level that produces pharmacological responses.FAAH inhibitors indirectly modulate the cannabinoid receptors (CB 1 and CB 2 ) and exhibit therapeutic effects with minimum CNS side effects, which are generally associated with direct agonists such as D 9 -tetrahydrocannabinol (THC) of the CB 1 receptor (Ahn et al., 2009;Leonard et al., 2017).
FAAH inhibitors decrease enzyme activity by reversible and irreversible inhibition binding modes.The FAAH-guided catalysis primarily involves a nucleophilic attack by OH of the Ser241 residue on the carbonyl group of the substrate, while Lys142 participates as an acid/base catalyst to activate nucleophilic Ser241.The Ser217 is also essential since its mutation to alanine lowers FAAH activity (Boger et al., 2000;Bracey et al., 2002;Patricelli & Cravatt, 1999).In recent years, many novel chemical templates have been explored to develop FAAH inhibitors.However, there remains a major challenge to identify the prototype FAAH inhibitor with desired drug-like characteristics of clinical relevance such as potency, selectivity stability and safety related to anandamide activity and other off-target effects (Bhuniya et al., 2019).More recently, FAAH inhibitor drugs have begun clinical testing for the treatment of various psychiatric disorders, including schizophrenia, posttraumatic stress disorder (PTSD), depression and neuropathic pain.Structures of some known FAAH inhibitors (Chobanian et al., 2014;Griebel et al., 2018;Keith et al., 2015;Long et al., 2009;Mor et al., 2004;Treijtel et al., 2019) are presented in Figure 1.
Guided by the ligand-based scaffold hopping strategy, we have previously identified an isatin-based lead FAAH inhibitor BSS-7 (Jaiswal et al., 2018) with an IC 50 value of 1.49 ± 0.03 mM (Figure 2).The potential activity of this compound was due to the presence of rigid benzimidazole moiety at C-3 of the isatin moiety.However, this compound was relatively larger and possessed a clogP of 5.15, and these characteristics are not ideal for a CNS-acting agent.Thus, to further optimize the lead compound, we designed and synthesized a set of structural analogs of BSS-7 by attempting various structural modifications (lead simplification) at the N-1 position and C-5/C-6 position of the isatin scaffold.The structural variations attempted on compound BSS-7 are depicted in Figure 2. In our earlier study, the incorporation of heterocyclic rings like thiazole and benzothiazole at C-3 of isatin scaffold resulted in poor or no FAAH inhibition activity (Jaiswal et al., 2018).Therefore, the FAAH active benzimidazole pharmacophore present in BSS-7 was retained in the further design of its structural analogs (compounds 6-13).
The bulky di-substituted benzoyl group at N-1 of BSS-7 was replaced with a relatively small, flexible allyl, propargyl or unsubstituted benzoyl moiety to obtain ligands possessing less rigidity and more flexibility (compounds 6-13).Further, a five-membered 1,3-dioxole ring was fused to the benzene ring of isatin (at R 2 ), and the benzimidazole fragment at C-3 of isatin was swapped with substituted aryl moieties resulting in a set of analogs (compounds 17-20) with variable lipophilicity.These structural modifications were attempted primarily to derive a comprehensive SAR for FAAH inhibition activity of isatin analogs.Besides focused variations were made at N-1 and C-3 positions of isatin scaffold to reduce the size and relatively decrease the rigidity and lipophilicity of a molecule that are essential indicators for potential FAAH-binding and blood-brain penetration properties, respectively.All these new analogs were evaluated for FAAH inhibition activity, and selected compounds were further evaluated for CNS properties, including neurotoxicity.

Chemistry
All chemicals were purchased from commercial suppliers and used without further purification.Reaction progress was monitored by TLC using silica gel 60 F254 with detection by UV cabinet.Melting points were determined in Pyrex capillary tubes using an Electrothermal 8103 apparatus and were uncorrected.All final compounds were purified by column chromatography.FTIR spectra were recorded on a Shimadzu FTIR 8400S infrared spectrophotometer using KBr pellets. 1 H NMR and 13 C NMR spectra were recorded on a Bruker 500 MHz spectrometer.Deuterated (d 6 ) DMSO was used for sample preparation of NMR.Splitting patterns of compounds are described as singlet (s), doublet (d), triplet (t), quartet (q), quintet (p) and broad (br); the value of chemical shifts (d) are given in ppm and coupling constants (J) in hertz (Hz).LC-MS spectra of synthesized compounds were obtained on a Shimadzu instrument.The negative/positive electrospray ionization method was applied in an LC-MS system to get the respective mass of analyzed compounds.The elemental analyses (CHN) data were obtained on Perkin Elmer 2400 CHN analyzer.

Synthesis of intermediates; 3chloroisonitrosoacetanilide, 4-trifluoromethyl isonitrosoacetanilide and 3,4methylenedioxyisonitrosoacetanilide (15)
About 0.054 moles of chloral hydrate in 50 mL of water was taken in a round-bottomed flask and sequentially, solutions of 0.01 mole of anhydrous sodium sulfate, 0.05 moles of substituted aniline in water and 0.0158 moles of hydroxylamine hydrochloride in water were added in this solution.The reaction mixture was refluxed over the water bath for about 40-45 min.Upon cooling the mixture, the solid product formed was extracted with ether.Ether was evaporated at room temperature, which yielded the crystals of respective isonitrosoacetanilide.The final product was recrystallized from ethanol (Hewawasam & Meanwell, 1994).

Synthesis of 5-trifluoromethylisatin (2b) and 5,6methylenedioxyisatin (16)
Concentrated sulfuric acid (5 mL) was warmed to 50 � C in a round-bottomed flask fitted with a mechanical stirrer.To this, 0.046 moles of dry substituted isonitrosoacetanilide were added slowly while maintaining the temperature between 60 and 70 � C.After adding isonitrosoacetanilide, the solution was heated up to 80 � C for 3-4 h.After completion of the reaction, the mixture was cooled to room temperature and poured into 10 to 12 times its volume of cracked ice, and kept it for 30 min.The resultant substituted isatin was filtered with suction, washed several times with cold water to remove the sulfuric acid, and then air-dried (Marvel and Hiers, 1925).

Synthesis of N-allyl and N-propargylisatin (4a and 4b)
To a solution of isatin or substituted isatin (0.01 moles) in DMF (50 mL) containing potassium carbonate (0.07 moles), allyl bromide, or propargyl bromide (0.02 moles) was added slowly with simultaneous stirring.The reaction mixture was then placed for stirring at room temperature for 24 h.After the completion of the reaction, contents were poured into the ice-cold water (500 mL), and the crystalline precipitate formed was filtered, dried and recrystallized from ethanol (Mandal et al., 2020).

Synthesis of N-benzoylisatin (4c)
Isatin (0.1 moles) was dissolved in 50 mL of DMF in a 250 mL round bottom flask and benzoyl chloride (0.1 moles) was added dropwise with constant stirring.Sodium hydride (0.001 moles) was added to the reaction mixture and refluxed for 5-6 h.After the completion of the reaction, the reaction mixture was cooled to room temperature, and the precipitate obtained was filtered, dried and recrystallized from ethanol (Jaiswal et al., 2018

General procedure for the synthesis of 5-nitroisatin (5)
Fuming nitric acid (1.0 mL, 0.035 moles) was added dropwise to a solution of indoline-2,3-dione (2.35 g, 0.05 moles) in sulfuric acid (10 mL) under 0 � C. The mixture was stirred for 5 h at 0 � C and then slowly poured into 500 mL of crushed ice.
The resultant yellow color precipitate was filtered, washed three times with water and then air-dried (Lahari & Sundararajan, 2020).

Synthesis of final compounds (6-13 and 17-20)
Equimolar quantities of substituted isatin and heteroaryl amine were dissolved separately in 10 mL of ethanol.Both solutions were mixed and acidified with a few drops of glacial acetic acid, and the reaction mixture was refluxed over a water bath for 12-24 h.The progress of the reaction was monitored by TLC, and the condensation reaction was carried out until a clear single spot was obtained.After the completion of the reaction, the contents were cooled to room temperature, poured onto a petri dish and allowed to stay overnight for crystallization.The following day, the crystals were collected, washed and recrystallized from ethanol (Jaiswal et al., 2018).

In vitro FAAH inhibition assay
In vitro FAAH enzyme inhibition assay was done by our earlier reported protocol (Jaiswal et al., 2018).The assay protocol followed is described in the supporting information, Section A.1.

Enzyme kinetics and reversibility studies
The enzyme reversibility and kinetics studies for both enzymes were done by earlier reported protocol (Jaiswal & Raja Ayyannan, 2021;Jaiswal et al., 2022).The detailed procedure is described in the supporting information, Section A.2.

Blood-brain permeability assay
A parallel artificial membrane permeation assay (PAMPA) was carried out to evaluate the blood-brain permeability of lead compound.Protocol validation and P e determination of reference and synthesized compounds were done by the reported protocol (Di et al., 2003;Mathew et al., 2016) mentioned in the supporting information, Section A.3.

Antioxidant assay
The antioxidant activity of synthesized compounds was determined by the DPPH assay with our previously reported method with minor modifications (Jaiswal & Raja Ayyannan, 2021).The details of the experiment are described in the supporting information, Section A.4.

Neurotoxicity studies
The most potent inhibitor 11 was screened for neurotoxicity using the rotarod apparatus at a dose of 60 and 30 mg/kg i.p. at 0.5, 1, 2 and 4 h time intervals to ascertain minimal motor impairment.The method was adopted from the reported procedure given by the National Institute of Health (Krall et al., 1978;Porter et al., 1984).The detailed assay procedure is described in the supporting information, Section A.5.

Molecular docking studies
Molecular docking studies of synthesized compounds against FAAH were conducted on a genetic algorithm-based computational package, AutoDock 4.2 (Gasteiger & Marsili, 1980).Ligand PF-750 co-crystallized with FAAH (PDB entry: 2VYA, resolution ¼ 2.75 Å) (Mileni et al., 2008) was docked for the validation of docking protocol.The docking run was carried out by the following three steps: (a) Preparation of protein, (b) Preparation of ligands and (c) Automated docking simulation.a. Preparation of protein: The X-ray crystal structure of humanized rat (h/r) FAAH complex with an inhibitor, PF-750 (PDB entry: 2VYA), was retrieved from the Protein Data Bank (https:/www.rscb.org/pdb).The 3D coordinate file was pre-processed using BIOVIA Discovery Studio 2019 to generate the monomeric chain, and the resulting monomeric structure was then cleaned by the removal of possible crystallographic artifacts.The cleaned monomeric FAAH protein was saved as a PDB file and used further for the docking simulation.b.Ligand preparation: Two-dimensional structure of all the test and reference compounds was drawn in ChemDraw Professional 15.0 and saved in PDB format by Chem3D 15.0 after energy minimization by the MM2 force field.c.Automated docking simulation: Firstly, the PDB protein file was loaded on AutoDock Tools (ADT; version 1.5.4) and the missing hydrogen atoms were added.Partial charges for proteins were added using Gasteiger-Marsili charges and Kollman charges, and nonpolar hydrogens were merged.The catalytic triad (Ser217-Ser241-Lys142) and Phe192, Tyr194 were assigned as flexible residues.
The resultant file was saved to pdbqt format.Further, all the energy-minimized PDB files of ligands (from step (b)) were loaded onto the AutoDock 4.

Free binding energy calculation
The free binding energy of all FAAH-inhibitor complexes was estimated by using Prime's molecular modeling and generalized Born surface area (MM/GBSA) module of Schrodinger suite 2021.All docked inhibitor-FAAH complexes were generated using Maestro pose viewer and uploaded in the Prime MM/GBSA module to obtain the results.The calculations were based on the implicit solvent model VSGB 2.0 implemented in Prime's MM/GBSA module (Li et al., 2011).

Molecular dynamics simulation studies
MD studies were performed on the most stable conformer of the enzyme-ligand complex obtained from the docking studies.Desmond module (Schrodinger Release 2020-3: Desmond Molecular Dynamics System, D. E. Shaw Research, New York, NY, 2020) was used for the simulation runs of the enzyme-ligand complex.The three complexes of FAAH protein bound with compound 11, compound 18 and reference ligand PF-750 were subjected to all atomistic molecular simulations run up to 50 ns.The employed simulation protocol involved three steps namely, system builder, energy minimization and molecular dynamics (Bowers et al., 2006).Initially, the system builder process enabled hydration of the protein-ligand complexes with a TIP3P solvent model in a cubic box size (10 � 10 � 10 Å) with a cutoff distance 5 Å.A force field OPLS3e was used for the simulation and the neutralization of the system was done by the addition of counter ions of Cl À .Further, the energy minimization of system was performed using the standard equilibration protocol implemented in Desmond, which contains several steps before the final production.Beginning with a Brownian dynamics NVT simulation for 50 ns at 300 K temperature, a simulation run was performed for the docked complex by keeping the trajectory with recording interval energy of 100 ps and energy of 9.6 to obtain the approximate number of 1000 frames (Bowers et al., 2006) The Maestro (Maestro, Schr€ odinger, LLC, New York, NY, 2020) graphic user interface (GUI) was used for the preparation of input files and visualization of the results.

In silico drug-likeness and ADMET prediction
For in silico drug-likeness and ADMET profiling, the SDF file of compounds was uploaded to the FAFDrugs4 online server (http://fafdrugs3.mti.univ-paris-diderot.fr/).Chemical structures SDF files of the synthesized compounds were generated using Instant JChem software (https://chemaxon.com/products/instant-jchem).Instant JChem is an OS-independent desktop application for scientists to manage and work with chemical structures and data using local and shared databases.
Further, isatin ( 3) was subjected to N-alkylation to generate the N-allyl, N-propargyl and N-benzoyl analogs (4a-4c).Final compounds (6-13, 17-20; Scheme 1 and 2) were synthesized via condensation of isatin and its fragments with 2-aminobenzimidazole or aromatic amine in the catalytic presence of glacial acetic acid and were obtained in good yields (50-90%).All final compounds were crystalline and displayed a sharp melting point range.All synthesized compounds were soluble in polar solvents except water.Moreover, the structures of the intermediates and the final compounds were confirmed by FTIR, 1 H NMR and 13 C NMR analysis.The details of spectral characterization are presented in the experimental section.

In vitro FAAH inhibition assay
The FAAH inhibitory activity profiling of the synthesized compounds was performed against the human recombinant FAAH enzyme using a well-established microplate-based fluorescent assay (Kage et al., 2007).The tested compounds showed inhibition against FAAH enzyme in nanomolar to micromolar range; 6.7 nM to 20.25 mM (Table 1).Reference FAAH inhibitor for in vitro assay. d Reference compound for docking study (Mor et al., 2004); nd: Not determined.2-fold more activity than the previously identified FAAH inhibitor BSS-7.
Based on the FAAH inhibitory data following SAR (Figure 3) is proposed for the tested compounds.Structural modifications attempted at two sites (R 1 and R 2 ) of BSS-7 resulted in a mixed inhibition profile with IC 50 values ranging from 6.7 nM to 20.25 mM.Among the analogs possessing variable substituents at R 1 (compounds 6-10) or R 2 (compounds 11-13), compounds 10, 11 and 13 have shown a better inhibition profile than the parent compound BSS-7.Except for compound 10, all N-1 (R 1 ) unsubstituted analogs (compounds 6-9) were significantly less active than BSS-7.Among the three N-substituted analogs 11-13, the flexible N-allyl bearing analog (compound 11) displayed the highest inhibitory activity, followed by N-benzoyl (compound 13) and N-propargyl (compound 12) analogs.These results suggest that the FAAH inhibition activity of compounds is primarily mediated by the presence of a benzimidazole ring at the C-3 imino terminal of the isatin scaffold, and by the presence of flexible and electron-rich allyl moiety at N-1 position (R 1 ).
On the other hand, the 1,3-dioxole ring fused analogs (compounds 17-20) possessing different electronegative substituents on the carbocyclic ring at C-3 displayed varying degree of FAAH inhibition activity.All the ring-functionalized analogs (compounds 18-20) were more active than the unfunctionalized analog (compound 17).The 4-fluoro substituted analog (compound 18) was �77-fold more active than the 4-chloro analog (compound 19), which in turn was � twofold more potent than its isomeric 3-chloro analog (compound 20).These results suggest that the degree of electronegativity of the halogen atom and its position modulate the FAAH inhibitory activity of these analogs.

Enzyme reversibility and kinetic studies
Further, the most potent inhibitor 11 was investigated for reversibility and kinetic studies to determine the binding mode of FAAH inhibition.The study was performed using earlier reported procedures (Jaiswal & Raja Ayyannan, 2021;Jaiswal et al., 2022).As shown in Figure 4A, compound 11 showed no time-dependent decrease in the rates of FAAHcatalyzed inhibition of 7-amino-4-methyl coumarin (AMC) that was pre-incubated with the enzyme for various periods (0, 15, 30, 45 and 60 min).Thus, it can be concluded that the tested compound 11 inhibits the FAAH enzyme reversibly.The reciprocal Lineweaver À Burk plots (Figure 4B) illustrate increased slope (unchanged V max ) and higher intercepts (K m ) with the increasing concentration of compound 11.The intersection point of the Lineweaver À Burk reciprocal plots was located in the second quadrant, which indicates that the inhibition mode of compound 11 was a competitive type of inhibition.
Further, to know the binding strength of compound 11 with the FAAH enzyme, the inhibition constant (K i ) was calculated (Figure 4C) by using Microsoft Excel 2013.From the Excel-generated Dixon plot between the reciprocal of the reaction rate (1/V) and the increased initial concentration of inhibitor (I, nM), the K i of lead FAAH inhibitor 11 was found to be 5 nM.

Blood-brain permeability assay
Blood-brain barrier (BBB) permeability is an essential criterion for any CNS drug.PAMPA was carried out to assess the blood-brain permeability (Pe) of lead compound 11 (Kansy et al., 1998, 2001).Assay protocol validation was performed by comparing the experimental permeability of seven commercial drugs reported by Di et al. (2003).The experimental and literature data showed a good correlation and are presented in the supporting information Table S01.PAMPA assay results showed that the lead compound 11 exhibited P e 10 À 6 cm s À 1 ¼ 5.25 ± 0.6, which indicates a high blood-brain permeation, as classified by Di et al. (2003).

Antioxidant activity
Oxidative stress plays a vital role in mood disorders and neurodegenerative diseases.The antioxidant molecule helps to counteract the unstable molecules that comprise free radicals by counteracting the adverse effects of oxidative stress.Isatin analogs are well reported, antioxidant agents (Shaikh et al., 2017).To know the antioxidant potential of compound 11, DPPH (2, 2-diphenyl-1-picrylhydrazyl) assay by our earlier reported protocol (Jaiswal & Raja Ayyannan, 2021).The results demonstrated that the compound 11 (100 lg/mL) possesses significant antioxidant activity (percent inhibition ¼ 65.2%) as compared to ascorbic acid (percent inhibition ¼ 61.2% (Tripathi & Ayyannan, 2020), and can be helpful in oxidative stress associate disease.

Neurotoxicity studies
Neurotoxicity is a major concern for the safety profile of CNS drugs.The most active compound 11 was evaluated for its neurotoxic properties by the rotarod test (Jaiswal et al., 2018).The test compound ( 11) was administered at dose 30 mg/kg (i.p) and 60 mg/kg (i.p), and the reference drug used was phenytoin (30 and 60 mg/kg, i.p).The responses (number of falls) were recorded after 0.5, 1, 2 and 4 h of dosing.Compound 11 was found safe for both doses and devoid of neurotoxicity (0/4 as shown in Table 2) after 4 h.Compound 11 showed the same efficacy as the reference drug.Results suggest that the lead compound 11 does not cause any motor impairment and thereby merits further development as a neuropathic pain agent.

Molecular docking studies of synthesized compounds 6-13 and 17-20
All compounds were subjected to molecular docking investigation to enumerate the binding efficacy and stability of the bound enzyme-inhibitor complex.The predicted free energy change for the protein-inhibitor interaction (DG) is further used to calculate the inhibition constant (K i ), and the results are presented in Table 1.Our docking protocol was validated by the superimposition of the re-docked PF-750 conformer Note: The data 0/4 indicates that there was no fall out of four animals tested.
to its reference conformer within the FAAH active site (Supporting information Figure S17), and the RMSD of docked FAAH-PF-750 was found to be 1.1 Å.
The free energy of binding data revealed that the N-benzoyl analog 13 showed greater binding affinity and interaction (DG ¼ À 10.15 kcal/mol; K i ¼ 36.34 nM) while the 5,6methylenedioxyisatin analog 20 (DG ¼ À 7.04 kcal/mol; K i ¼ 6870 nM) showed weak affinity and interaction within the active site of FAAH.The specific interactions observed between the ligands and the active residues of FAAH are presented in Table 3.There was a poor correlation between the experimental IC 50 values and the predicted binding energies as the most potent FAAH inhibitor 11 (6.7 nM) ranked 6th in the docking studies, and the least active inhibitor (20.25 mM) ranked 4th in the in silico screening.However, the virtual binding data (DG) showed a linear relationship to the predicted inhibition constant (K i ).
Visual inspection of the binding mode of all inhibitors (Figure 5) within the active site of FAAH enzyme (PDB ID: 2VYA) was done to identify the binding orientation and interactions that contribute to the stability of the enzymeinhibitor complex.All inhibitors shared a common binding mode and were positioned at the center of the FAAH active site (Figure 5A, 6-13 and B, 17-20).Further investigation of interactions of inhibitors with the anchoring amino acid residues of FAAH led to the following observations.The inhibitors were found to interact with residues Ser217, Ser241, Phe192, Ile238, Gly239, Val491, Tyr194 and Trp531, and the binding pattern was very much similar to that of the reference FAAH inhibitor PF-750, and the previously identified FAAH inhibitor BSS-7.These findings suggest that the new isatin-based rigid analogs possess the requisite pharmacophoric features to bind efficiently with the FAAH enzyme.In all compounds, the isatin nucleus occupied the center of the cavity while the benzimidazole or other heterocyclic ring is oriented towards the catalytic triad region lined by Ser241, Ser217 and Lys142 residues.
Hydrogen bond interactions were observed between all the docked molecules with one or more active site residues of FAAH, and these interactions possibly stabilized the enzyme-inhibitor complex.Hydrogen bonding between the H atom of NH of Tyr194 and carbonyl oxygen (2-C ¼ O) of isatin was observed in compounds 6 and 8; while the H atom of NH of Ser241, Ser217 and Gly239 showed H-bond interaction with compounds 7, 11 and 10, respectively.Compound 9 showed H-bond interactions with the H atom of NH of Ile238, Gly239 and Ser241 through its carbonyl oxygen.Whereas compounds 7, 8 and 12 offered hydrogen bonding with Ser241, Pro484 and Gly485 residues, respectively, through the N atom of benzimidazole ring.In addition, p-p interaction was observed between Phe192 and benzimidazole moiety in compounds 6, 8 and 13, while these compounds showed pi-sigma interaction with Ser193, Leu404 and Thr488, respectively.The benzodioxole bearing analog 18 besides showing approximately a 90-degree deflection in the orientation than compound 11, showed H-bond interaction through an oxygen atom of benzodioxole ring with the H atom of Ile238.The protein-ligand interactions and their distances observed in the virtual FAAH-inhibitor complexes are presented in Table 3.
Further, examination of the binding conformation of compound 11 (Figure 4D) led to the following observations.The indoline backbone occupying the center of the cavity is oriented towards the catalytic triad region lined by Ser217, Lys142 and Ser241 residues.The main factors imparted stability to the enzyme-inhibitor complex were p-alkyl, p-amide stacking and hydrogen bonding interactions.In the complex, the benzene ring of isatin moiety was found to show p-alkyl interaction with Phe192 and Ile238 located near the central cavity.Pi-alkyl interaction was also observed between Val491 and benzimidazole ring.Besides this, the five-membered ring of isatin moiety, at 5.2 Å distance, showed p-amide stacking with carbon and oxygen atom of Ile238 and N atom of Gly239 located in the oxyanion hole.In addition, H-bond interaction between the oxygen atom of isatin moiety and the H atom of NH of Ser217 in the catalytic triad was also observed.These interactions altogether resulted in the firmness of the virtual FAAH-inhibitor 11 complex.
A comparison of the binding mode of the lead inhibitor 11 with the reference inhibitors PF-750, BSS-7 and JZL195 resulted in the following observations.Compound 11 showed a similar orientation as that of JZL195 (shown in Figure 4C) and they shared some common interactions, viz., H-bond interaction with catalytic site residue Ser217 and p-alkyl interaction with Val491.Compared to the virtually predicted bioactive conformer of BSS-7, the lead inhibitor 11 exhibited a 90-degree flip in the orientation, allowing the flexible allyl tail (at N-1) to orient towards the catalytic triad, and directing the benzimidazole moiety (at C-3) to the cavity lined by Tyr194, Thr488 and Val491 residues.This preferred orientation, along with p-alkyl and hydrogen bonding interactions imparted by the energetically favored conformer of the lead compound 11, advocates for its higher potency (�1400 times) compared to BSS-7 and other tested ligands.

Binding free energy calculation
Since the virtual docking study cannot estimate the binding free energy of docked complexes, the binding free energy was calculated to assess the stability of the virtual  complexes.Prime's molecular modeling and generalized Born surface area (MM/GBSA) approach based on the generalized Born model and solvent accessibility method was used for the calculation of binding free energy and the obtained results are presented in Table 1.The binding free energies were observed in the range of À 47.75 (compound 18) to À 62.08 (compound 13) kcal/mol.

RMSD analysis
Initially, protein-ligand RMSD was calculated to find out the average change in displacement of a selection of atoms from  the initial structure.The outcome demonstrates that the docked complex showed an RMSD value in the acceptance range, that is, 1-3 Å (Figure 6).Compounds 11 and 18 (0.4-1.6 Å) demonstrated better protein-ligand stability than the reference inhibitor (0.4-2 Å).After 30 ns, the highest fluctuation was observed in the reference inhibitor PF-750 (Figure 6A) and lead compound 11.However, it was stable after 40 ns in compound 11 (Figure 6B).Compound 18 showed the least or minimum fluctuation during the simulation period of 50 ns (Figure 6C).The heavy atoms of each ligand were also checked for the RMSD value to observe the stability throughout the simulation period.The observed results suggest the finest structure reconstruction of tested complexes during the simulation.

RMSF analysis
The amino acid fluctuation was calculated by RMSF.The average RMSF values for native protein for all three complexes (PF-750, compound 11 and compound 18) were found to be 2.5 Å. Increased RMSF was observed in tested ligands in the region of 60 to 80 residues (initial loop Figure 7A, B).The protein active site regions Lys142, Ser217, Ser241, Gly240 and Ile238 have shown good stability, as the middle loop region exhibited lower RMSF value (Figure 7).The ligand RMSF value also indicates docked ligands' stability during the entire 50 ns simulation (Supporting information, Figure S17).

Protein-ligand interactions
The simulation results suggest that the docked complex is stabilized primarily by the interactions of compound 11 with the dynamic gate residues (Phe192, Tyr194) of FAAH.
Compound 11 interacts with FAAH active site residues Phe192 and Tyr194 mainly through hydrogen bonding (Figure 8A).Further, hydrophobic interaction with Tyr194, Leu380, Phe432, Val491, Leu401 and Met495 (Figure 8B) and water bridges with residues Thr488 and Ser193 play a supporting role in ligand binding.The timeline interaction plot of various types of interaction for 50 ns runtime (Figure 8C) indicates that a minimum of four contacts was present throughout the simulation period, accounting for the greater stability of the inhibitor-FAAH complex.
Furthermore, MD simulation studies on the least stable FAAH-inhibitor 18 complex revealed that compound 18 showed a relatively minimal number of contact points (Figure 9B, C) with active site residues compared to lead compound 11.This accounts for the compound's (18) poor binding energies and stability.Compared to the highly stable inhibitor 11, compound 18 lacked hydrogen bonding interaction with the dynamic gate residues Phe192 and Tyr194, which are crucial for the activity of FAAH.The 50 ns simulation data suggests that the stability of compound 18 was mainly driven by hydrophobic contacts with residues; Phe192, Tyr194, Phe244, Phe382, Phe388, Leu404, Ile405, Val491 and Met495 and water bridges with residues Met191, Ser217, Ile238, Ser241, Ser376, Thr377, Thr488, Leu401 and Pro484.

Pose analysis after MD simulation
The MAGL-inhibitor (compounds 11 and 18) complexes were analyzed for conformational changes after MD simulation.The simulation results demonstrated no conformational change in the inhibitor molecules compared to their docking conformers.These results also indicate our docking studies' reliability.However, additional/new binding interactions have been observed after the simulation compared to the docking conformers (Figure 10).After simulation, compound 11 showed the hydrogen bond interaction with Phe192.The benzimidazole ring displayed pi-sulfur interaction with Met495, and the five-member ring of the isatin nucleus showed pi-sigma interactions with Leu404 and Thr488.These additional interactions indicate the greater flexibility of compound 11 for the FAAH active site and maybe the possible reason for greater FAAH inhibitory activity.
As shown in Table 4, all synthesized compounds followed Lipinski's rule of five.Their log P values were � 4, which is an ideal characteristic for CNS-targeted drugs.Drugs with log S values less than 0 and greater than À 4 are well absorbed by blood and tissue.All synthesized compounds displayed a reduced water solubility index, suggesting that the compounds are highly lipophilic and thus possess the ideal characteristics of CNS-targeted drugs.
In the ADMET profiling studies (Table 5), it was observed that all compounds showed good oral bioavailability according to Veber's and Egan rule which state that compounds with rotatable bonds �10; TPSA � 140 Å; H-bond donors þ H-bond acceptors � 12 and (0 � TPSA � 132) and (À 1 � logP � 6) are suitable for oral bioavailability respectively.All compounds were found to be non-inducer of phospholipidosis which means compounds were not hepatotoxic and had acceptable metabolic stability.For drug safety concerns, compounds are screened through the GSK 4/400 rule and Pfizer 3/75 rule, which states that compounds with a logP less than 4 and an MW less than 400 Da may have a more favorable ADMET profile and compounds with a high log P (>3) and low TPSA (<75) are approximately 2.5 times more likely to be toxic as to be clean, respectively.All compounds passed the GSK 4/400 rule, while Pfizer's 3/75 rule showed a warning about drug safety.While consideration of all parameters of the FAF-Drug4 web server, all compounds studied got accepted status, as shown in Table 5.From these results, it can be concluded that all compounds can be administered through the oral route, which is a preferable route for any drug candidates under development.

Conclusion
In summary, a set of structural analogs of previously known FAAH inhibitor BSS-7 was investigated for their in vitro and in silico FAAH inhibitory properties.The in vitro FAAH inhibition assay identified an N-allylisatin analog 11 possessing nanomolar inhibitory activity (6.7 nM) with a reversible and competitive mode of inhibition.In silico molecular docking studies and assessment of the stability of the docked FAAHinhibitor complexes by 50 ns MD simulation run provided insights on the crucial binding interactions and specific contact points between the inhibitor and FAAH active site residues.Furthermore, the lead FAAH inhibitor 11 exhibited good blood-brain permeation and antioxidant profiles.In silico exploration of drug-like and ADMET properties suggests that the lead compound possesses desirable pharmacokinetic properties with an acceptable safety profile.Thus, the present study, guided by the lead simplification strategy, yielded an improved and highly potent non-neurotoxic FAAH inhibitor compound 11 and is thus considered for further evaluation.

Figure 4 .
Figure 4. (A) Reversibility study of compound 11 by jump dilution assay; (B) Lineweaver-Burk plot of compound 11 with FAAH catalyzed inhibition (1/V) of AMC, in the absence (control) and presence of various concentrations of substrate (1/[S]).(C) Dixon plot showing the K i value of lead FAAH inhibitor 11.

Figure 8 .
Figure 8. Protein-ligand interactions of compound 11 in FAAH active site during 50 ns simulation: (A) Percentage interaction of the ligand with protein, the florescent green area represents the hydrophobic interaction, and solvent exposure area is highlighted in grey color; (B) Stacked bar showing different types of protein-ligand contacts; (C) Timeline representation of protein-ligand contact for a total simulation run time of 50 ns; the top panel shows the total number of specific contacts the protein makes with the ligand throughout the trajectory and the bottom panel shows which residues interact with the ligand in each trajectory frame.

Figure 9 .
Figure 9. Molecular dynamics simulation data of docked complex of least active inhibitor 18-FAAH: (A) Percentage interaction of the ligand with protein; florescent green labels represent the interacting amino acid residues, green line represents pi-pi stacking and solvent exposure area is highlighted by grey color; (B) Stacked bar showing various protein-ligand contacts; C) Timeline representation of protein-ligand contact for a total simulation run time of 50 ns; the top panel shows the total number of specific contacts the protein makes with the ligand throughout the trajectory, and the bottom panel shows which residues interact with the ligand in each trajectory frame.
Further, molecular dynamics (MD) simulation studies were performed to assert the docked compound's stability within the active site of FAAH.The enzyme-inhibitor complexes of the most active compound 11, least active compound 18 and reference inhibitor PF-750 were subjected to MD simulations up to 50 ns by using the Desmond program on a Schrodinger platform (Schrodinger Release 2020-4: Desmond Molecular Dynamics System, D. E. Shaw Research, New York, NY, 2020.Maestro-Desmond Interoperability Tools, Schr€ odinger, New York, NY, 2020).The preparation of input files and visualization of results was done by Maestro graphic user interface (Maestro, Schr€ odinger, LLC, New York, NY, 2020).The protein-ligand complexes of compounds 11, 18 and reference inhibitor PF-750 were observed for root-mean-square deviation (RMSD), root mean square fluctuation (RMSF), protein-ligand contacts and other ligand properties like the radius of gyration (RG), solvent accessible area (SASA), etc.

3,4-Methylenedioxyisonitrosoacetanilide
(15): Brownblackish crystal; yield 90%; m.p. 160-162 � C, R f 0.85; FTIR (KBr) m max used for molecular docking studies are 80 � 60 � 60 with a line spacing 0.375.The grid center coordinates were situated at x:46.172; y: À 17.338; z:  À 38.592.The output of the grid was saved in GPF file.We also prepared the DPF file of the ligands.After that, the AutoGrid and AutoDock module was run in the command prompt, which gave the conformer log dlg file.The generated conformers of each ligand were then written in PDB format and analyzed in Discovery Studio 2019.
2 program and flexible torsions were assigned with the AutoDock module, and the acyclic dihedral angles were allowed to rotate freely.The file was then converted into pdbqt file format.After the preparation of both pdbqt files, we generated the grid box, the specific region of binding of the ligand in protein.The FAAH grid was centered between the oxyanion hole and catalytic grid, and the xyz coordinates were

Table 4 .
Predicted physicochemical descriptors of synthesized compounds 6

Table 5 .
In silico ADMET profiling of synthesized compounds 6-13 and 17-20.number of stereo centers and Fsp3.Whereas oral bioavailability and drug safety profiling it is based on several rules like the Veber rule