Computational analysis of eugenol inhibitory activity in lipoxygenase and cyclooxygenase pathways

Chronic inflammation is triggered by numerous diseases such as osteoarthritis, Crohn's disease and cancer. The control of the pro-inflammatory process can prevent, mitigate and/or inhibit the evolution of these diseases. Therefore, anti-inflammatory drugs have been studied as possible compounds to act in these diseases. This paper proposes a computational analysis of eugenol in relation to aspirin and diclofenac and analyzing the ADMET profile and interactions with COX-2 and 5-LOX enzymes, important enzymes in the signaling pathway of pro-inflammatory processes. Through the analysis of ADMET in silico, it was found that the pharmacokinetic results of eugenol are similar to NSAIDs, such as diclofenac and aspirin. Bioinformatics analysis using coupling tests showed that eugenol can bind to COX-2 and 5-LOX. These results corroborate with different findings in the literature that demonstrate anti-inflammatory activity with less gastric irritation, bleeding and ulcerogenic side effects of eugenol. The results of bioinformatics reinforce studies that try to propose eugenol as an anti-inflammatory compound that can act in the COX-2/5-LOX pathways, replacing some NSAIDs in different diseases.


Methods
protein and ligands structural data. Crystal structures were downloaded as .pdb files from Protein Databank Website (www.rcsb.org). The initial biopolymers were simplified by deleting all ligands and water in the structures using Discovery Studio (version 2016). It was selected and chosen and prepared one of the chains (chein A) of each protein for the docking analyze. The following Crystal structure access codes from Protein Databank (www.rcsb.org) were downloaded and employed for the docking study: COX-1 (3N8W, 3N8X, 3N8Y), COX-2 (3LN1, 1CVU, 4OJT), and 5-LOX (3O8Y, 3V99). The Ligand 3D structures were downloader as .sdf files format from Zinc Library (https ://zinc1 5.docki ng.org/subst ances ) and optimized by PyMOL (version 2.1.1) and saved as .PDB files. The protein crystal structures were optimized using PyMOL free software and Discovery Studio (version 2016) and saved as .PDB files. All prepare for the molecular docking studies were carried out using AutoDock Vina (version 1.1.2) in CHIMERA (version 1.12).
Docking. The molecular docking technique was used in order to predict the binding geometry requirements of the target molecules to predict the enzymatic mechanisms interectios os NSAIDs and eugenol wich COX in anti-inflammatory processes and in antitumor activity. Eugenol, diclofenac and aspirin interactions with the COX-1, COX-2, and 5-LOX enzymes was performed by molecular docking, the compound was docked using SwissDock (https ://swiss dock.ch), a web tool for rapid ligand-based virtual screening of small to unprecedented ultra-large libraries of small molecules 45 . After docking completion, ligand conformations displaying greatest binding affinity and lowest docked energies were chosen and re-docking on Autodock Vina 46 , using Autodock Vina docking protocol 47 . The hydrogen bonds, bond lengths and hydrophobic interactions between enzyme (COX-2) and all ligands were determined by using PyMol. Chimera, PyMol and Discovery Studio programs also ware used for visual inspection and graphical representations of the docking results.
Docking validation. All the crystalline structures analyzed were obtained by the X-ray diffraction method and presented resolutions smaller than 2.8 Å (Table 1). How validation of the docking process it was performed using Root-Mean-Square-Deviation (RMSD) calculations, calculated using the web-based Dincdocking (https ://dinc.kavra kilab .org). RMSD values were calculated by comparing the lowest-energy conformation with each fragment conformation, considering all heavy atoms of the ligand. For the validation of the enticer binding with the amino acid fragments, only connections with up to 3.0 Å of compliance was considered and analyzed. All values and links analyzed and described were exhaustively compared to values and data described in the literature, especially those described and related to the structures searched in the Protein Databank website (www. preADMet. The pharmacokinetic and pharmacodynamic properties of ADMET, such as absorption in the human intestine-HIA, plasma protein binding-PPB and blood-brain partition coefficient (log BB) were analyzed by in silico studies by PreADMET (https ://pread met.bmdrc .kr) and SwissADME (https ://www.swiss adme. ch) online databases to evaluate the pharmacokinetic parameters 50 to relate drug absorption, metabolism and toxicity 51 for the drugs and eugenol. Prior to that, SDF (Structure Data File) and SMILES (simplified molecular input line entry system) strings were utilized throughout the genera-tion process. The results were analyzed and compared.
Dinamics. The geometries and energies of the excited states were calculated using the Time Dependent Density Functional Theory (TDDFT), using IQMol (version 2.11.1) software to perform computational calculations 52,53 . The B3LYP functional was used to predict the structural properties as well as the excitations involving charge transfer 54 . The choice of B3LYP is due to the fact that this base to be considered to be universally functional and presents excellent performance when applied to a variety of systems at a relatively low computational cost 55 .

Results and discussion
Evaluation of the preADME profile of the eugenol and NSAIDs (diclofenac and aspirin). Compared to the NSAIDs, from the druglikeness in silico studies performed (Table 2), eugenol was observed, exposed to drug dispensers, because they were in accordance with the criteria of noncompliance with CMC rule, Rule of Five, and the leadlike rule, it may be in accordance with these indices to be qualified as druglikeness compound [56][57][58][59][60][61][62][63] . Eugenol exhibited very similar behavior to diclofenac and aspirin. Molecules that no violate these rules like eugenol don't have problems of bioavailability. The Eugenol violates the MDDR rule in two parameters: (1) the number of ring bonds and (2) as to the number of rotation bonds, due to the number of violations is classified as druglike and nodruglike mean structure 63,64 . Interactions involving aromatic rings are major contributors to protein-ligand recognition and concomitantly to drug design. While that reducing the number of aromatic rings of a molecule might improve its physicochemical properties, such as solubility increasing its bioavailability [63][64][65] . On the other hand, leadlikes are generally smaller molecules such as eugenol and allow structural incorporations to increase effectiveness during lead optimization and can be incorporated as lead optimization processes 66 . Druglikeness values of the substances are very similar, especially between eugenol and aspirin. Probably because they have more similar chemical structures. These values demonstrate that eugenol has a chemical structure druglikeness. Table 3 shows a comparison of the pharmacokinetic and physicochemical properties of eugenol, diclofenac and aspirin. Biological compounds that are delivered now, need to cross the intestinal barrier to reach their pharmacological target. Eugenol has better intestinal absorption (HIA) ~ 96.8% than other compounds. The permeability of Caco-2 and MDCK cells is more effective for eugenol when compared to diclofenac and aspirin [67][68][69] . The substances exhibit many pharmacodynamic similarities. It is worth mentioning the greater capacity of eugenol to cross the blood brain barrier, which is probably related to its analgesic action, as well as its better intestinal absorption and permeability in Caco-2 and MDCK cells.
In silico analyzes showed that eugenol has a greater capacity for binding to plasma proteins, justifying a possible plasma transport mechanism and greater capacity for transposing the blood-brain barrier, when compared to diclofenac and aspirin [67][68][69] . As for solubility in water, eugenol is practically insoluble in water, 5.2 × 10 -3 mol/L, as well as 6.3 × 10 -3 mol/L in a buffered solution (pH 7.4). The physicochemical parameters such as SKlogP related to the lipophilicity of chemical substances show the compound is quite liposoluble [70][71][72] . That indicates the eugenol availability in the plasma membrane when absorbed by the organism (Table 3 and Supplement 1). The structural similarity between the molecules of the substances is evidenced especially between eugenol and aspirin especially by the comparison between the number of rotatable bonds, H-bond acceptors number, heavy atoms number and molecular mass (Fig. 1). In silico toxicity tests (Table 4) predicted similar toxicity for the Ames test for the tree compounds. For TA100 and TA1535 two strains of Salmonella typhimurium that are frequently used in the Ames test containing the same base pair replacement mutation hisG46 [73][74][75] , eugenol exhibit aspirin-like behavior. The model predicted positive carcinogenicity results for eugenol in both rats and mice, differing from the other two compounds that show negative results, except for aspirin in rat tests. All the compounds under study also present a medium risk for inhibition of the hERG gene, encoding the rectified potassium channel of the rectified voltage into the heart (IKr) involved in cardiac repolarization. The results point to the insured use of eugenol when compared to the other drugs analysed. The substances exhibit many toxicological similarities. This similarity is even greater between eugenol and aspirin. Probably because they have more similar chemical structures.
COX-2-specific NSAIDs are weak organic acids, and lipophilic 24,28,41,76 . Thus, the lower the pH, the greater is their lipophilicity like eugenol. This combination of chemical properties allows that similar NSAIDs COX-2-specific compounds like eugenol (as well as conventional NSAIDs) to cross lipid membranes, including the blood-brain barrier, and to accumulate in acidic tissues such as the stomach, renal medulla, and sites of inflammation 77 .

Evaluation of the docking of COX with eugenol and NSAIDs (diclofenac and aspirin). COX
isoforms have two active sites, cyclooxygenase and heme-dependent peroxidase, which catalyze the conversion of AA to prostaglandin G2 (PGG2) and the conversion thereof to PGH2, respectively (Figs. 2, Supplements 2, 3). On the opposite side of the protein from the membrane-binding domain, the peroxidase active site consists of the heme positioned at the bottom of a shallow cleft 78 . The structural modification results of an exchange of valine at the position of 523 in COX-2 for relatively bulky isoleucine (Ile) residue in COX-1 at the same position of the active site of the enzyme 41,79 resulting in a cavit there is not observed in COX-1 that results in steric hindrance at the active site of the enzyme, constraining the valine fragment in COX-2, which is much less bulky. This larger opening of the active site of COX-2 allows greater accessibility of larger molecules that could not interact with the active site of the enzyme. Which makes the COX-2 isozyme active site (volume = 394 Å 3 ) is about 25% larger than the COX-1 isozyme-binding site (volume = 316 Å 3 ). Table 3. Physicochemical properties and ADME values to the substances analyzed by the preADMET and SwissADME.  (C) Molecular Structure 2D and 3D of Aspirin. 2D structure were building in the software ACD Labs ChemSketch release 12.0 (https ://www.acdla bs.com/resou rces/freew are/chems ketch /). 3D structures were downloader from Zinc Library (https ://zinc1 5.docki ng.org/subst ances ) and saved as .sdf files. The structures were optimised using IQMol software, version 2.11.1 (https ://iqmol .org/). www.nature.com/scientificreports/ There are several common features between COX-1 and COX-2 active sites, the main hydrophobic channel, the catalytic Ser-530 and the mouth having polar residues such as Arg-120. However, COX-1 is characterized by the absence of the side pocket and by a narrower main hydrophobic channel 44 . The COX-2 active pocket (Table 5 COX-2 selective inhibitors explicitly bind to this secondary-binding active site pocket lined resulting in the specific inhibition of COX-2 activity. In a recent mutational study described the involvement of hydrophobic pocket residues in the proper positioning of fatty acid substrates for oxygenation evidencing that highly potent and selective COX-2 inhibitors should possess a pharmacophore which can selectively bind in the secondary pocket and deliver sufficient steric bulk to block the hydrophobic channel of COX-2 81 . Diclofenac (2-[2-(2,6-dichloroanilino) phenyl] acetic acid) is NSAIDs COX-2 non selective inhibitor employed for the treatment of inflammatory symptoms (pain, redness, heat, and swelling) by means of block the formation of prostaglandins resulting in the reduction of acute and chronic inflammation [82][83][84][85] .
The computational data showed (Supplement 2B,F) the docking of diclofenac in the COX-2 shows interactions by H-bond with the cyclooxygenase active pocket catalytic residues Tyr 385 Ser 530 and 385; π-π T-shaped interactions between Phe 518, Trp 387 residues and non-chlorinated aromatic ring. There were observed alkyl and π-alkyl bonds between diclofenac molecule and Val 523, Val 349, Ala 527, Leu 352 amino acid fragments, interactions there is interaction of π-electron cloud over an aromatic group and electron group of any alkyl group. The residues Tyr 355, Tyr 348, Ser 353, Phe381, Phe 205, Gly 526 and Met 522 interact with diclofenac molecule by Van der Walls forces. All these interactions allow the diclofenac molecule to act by blocking the canal by an esoteric hindrance, causing its narrowing to impede the access of arachidonic acid to the active site where its catalytic conversion to prostaglandin occurs.
Aspirin (2-Acetoxybenzoic acid) is another nonsteroidal anti-inflammatory drugs target the cyclooxygenase enzymes (COX-1 and COX-2) to block the formation of prostaglandins resulting in the reduction of acute and chronic inflammation 86,87 . Similarly to the data found for diclofenac, aspirin (Supplement 2C,G) demonstrates its action by blocking the hydrophobic channel in the cyclooxygenase active pocket. The molecular docking of  www.nature.com/scientificreports/ Eugenol similarly to diclofenac and aspirin blocks the cyclooxygenase hydrophobic channel of the enzyme by means of hydrophobic H-bonds interactions with the fragment Ser 530, and hydrophobics interactions with Tyr 385. Eugenol, diclofenac and aspirin demonstrates the ability to interact strongly with the amino acid Ser-530, and the mouth has polar residues like Tyr 385 through hydrophobic interactions. Ser-530 and Tyr-385 are important for the inhibition of COX-2 by several compounds besides aspirin 90 . Ser-530 has also been shown to influence the stereochemistry for the addition of oxygen to the prostaglandin product. The catalytic residue Tyr385 is responsible for the conversion of AA to prostaglandin G2 by the transfer of an electron to the heme from Tyr-385 of the protein generates to tyrosyl radical in the cyclooxygenase active site.
Eugenol has also been shown to inhibit arachidonic acid catalysis in prostaglandin G2 directly in the active pocket site at the end of the hydrophobic channel of the COX by H-bond interaction with Ser 530. In the same pocket interact by hydrophobic forces with the fragments Ser 353, Phe 381, Phe 518, Gly 526, Trp 387, Tyr 348, Tyr 385, alkyl interaction with Val 349, π-alkyl with Val 523 and π-sigma with Leu 352 (Fig. 2 and Table 5).
Diclofenac, aspirin and eugenol has been shown to inhibit arachidonic acid catalysis in prostaglandin G2 directly in the active pocket site at the end of the hydrophobic channel of the COX by polar interaction with Ser 530 (Supplement 2D,H). Inhibition of the active binding site can result in specific inhibition of COX-2 activity (Supplement 3C,F).
Charges of atoms (eV), distribution of relative electron density, vibration energy, localization and energy of highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) Hartree Energy, molecular electronic density are very strong similarity between those molecules (Supplement 4) particularly the gap energies HOMO-LUMO between eugenol and aspirin (3,769 eV e 3,612 eV respectively) as shown in Figs. 3A. Those molecules also have a very close surface area, eugenol has a surface area of 388,859 Å 2 and aspirin 379,426 Å 2 . That explain them similar behaviour and interactions in the COX-2 hydrophobic active pocket.
Eugenol also occupies an orientation very similar to the binding mode of arachidonic acid in the COX-2 active site. The computational results evidence this similar orientation (Supplements 2, 3). According to the computational analysis, eugenol interacts with regions of the active site of COX-2 into the hydrophobic pocket. Ser 530 Table 5. Interaction types and amino acids involved in the inhibition of COX-2 cyclooxygenase catalytic (PDB ID: 4OTJ) with NSAIDs and eugenol.

Hydro-phobic interaction
Pi-sigma/Pi-sulfur/ Pi-amide interaction www.nature.com/scientificreports/ and Tyr 385 are the key amino-acid fragments that contribute considerably in the inhibition of the protein by the interaction protein-ligand of the active pocket of the enzyme. Eugenol has also been shown to act on the inhibition of the active site in which peroxidase is responsible for converting prostaglandin to G2 into prostaglandin H2 by the interaction with the group with the lateral methyl radicals of the pyrrole rings of the heme group. The compound also acts on the inhibition of prostaglandin G2 peroxidation in prostaglandin H2 interacting with the carbonic chain of the branching of the pyrrole rings that present the carboxyl function by salt bridge whit hem group, salt bridges contribute little to protein stability but, can make crucial contributions to ligand-protein binding stability. Eugenol also interacts with peroxidase active pocket fragments by van der Waals interaction with Ala 274, Arg 222, Gln 289, His 2014, His 207, Ile 274, Lys 211 and Thr 212; Leu 294 π-alkyl interactions and π-sigma whit Val 291 (Fig. 2).

Pi-Pi T-shaped interaction
The peroxidase site presents a heme group like cofactor. The heme cofactor is not bound covalently to PGHS and there are relatively few protein-heme interactions by Van der Waals interactions. The heme iron is coordinated on the proximal side with a nitrogen of a His 207. The coordination with a His 207 is conserved across the heme dependent peroxidases, however, in this peroxidase, the bond length is much longer than normal. A relationship between heme-nitrogen bond suggesting that interactions on the distal face of the heme affect the proximal bond length and reduction potential of PGHS.
The interaction of the binders with the structures obtained in the PDB found by the algorithm was analyzed using the Root-Mean-Square-Deviation (RMSD) calculation, which generally measures the degree of similarity between the structures. RMSD values < 2.0 Å (Fig. 3B) and the low ΔG binding values (Tables 6 and 7) indicate topological similarity between structure and the cavity in the active site and evidence the binding capacity of the ligand in COX-2.
Both the cyclooxygenase and the peroxidase active sites are located in the catalytic domain of the COX-1/2 isoforms. Docking analysis of eugenol with the COX-1/2 isoforms (Table 6) shows the eugenol's ability to interact with both isoforms, however a slightly larger docking binding energy module of the protein-ligand complex is observed with the COX-2 cyclooxygenase (− 6.69 kcal/mol) compared to the complex formed with COX-1 (− 6.20 kcal/mol). Eugenol also have showed the ability to interact with the active peroxidase site (− 6.87 kcal/ mol), and present a higher inhibition constant against the active site of peroxidase. The large number of NSAIDs and COX-2 selective inhibitors bind in cyclooxygenase active site but not in the peroxidase site 91 . Only Few COX inhibitors have been capacity against the peroxidase activity. Inhibited COX-2 cyclooxygenase, for example, in still capable to accommodate large hydroperoxide substrates such as PGG2 92 . The mean binding energies of the ligands with COX-2 are fairly close values. However, eugenol showed potential to inhibits both active sites of COX-2, increasing the potential of enzymatic inhibition to COX-2 activity of the eugenol wich a total docnkin complex energy of -13.56 kcal/mol highest the docking energy of Celecobix, a selective COX-2 inhibitor.
The binding forces of the compounds with COX-2 are also very close among the analyzed substances (Table 6). Arachidonic acid has a higher binding energy with the Heme group and the opposite side. However, eugenol has binding energy very close to aspirin. These results suggest that eugenol may interact with the Heme group and the opposite side, inhibiting the COX-2 enzyme, similarly to aspirin. In parallel, the weaker interactions between eugenol and COX-2 suggest that the inhibitory action can be reversed. This possibility can be used for studies that involve reversing side effects caused by continuous COX-2 inhibition in chronic patients who need to make continuous use of NSAIDs. Table 7 shows the interaction of eugenol with 5-LOX binding sites. The binding energy of eugenol with the active sites of 5-LOX, suggests the interaction between eugenol and 5-LOX. This factor may be an indication for   93,94 . Dual inhibition of COX-2 and 5-LOX enzymes of the arachidonic pathway is a evidence to the capable of the eugenol as anti-inflammatory agent action 95 . In vitro studies suggest that eugenol inhibits 5-LOX by a non-competitive mechanism, decreasing the production of C4 leukotriene in Polymorphonuclear leukocytes 94 . Therefore, the computational results of the molecular interaction of eugenol with 5-LOX reinforce the evidence from in vitro studies. Dual COX-2/5-LOX inhibitors have the advantages of enhanced anti-inflammatory potency with better safety profile. Whereas the traditional NSAIDs, like aspirin and diclofenac, inhibit cyclooxygenase pathway non-selectively and produce gastric mucosal iritation and 'Coxibs' which are selective COX-2 inhibitors, cause adverse cardiovascular events 37 . 5-LOX/COX double inhibitors have been the potential to safety and efficacy to treat inflammation processes, blocking the formation of the prostaglandins and leukotrienes 92 , and could provide numerous therapeutic advantages in terms of anti-inflammatory activity, improved gastric protection and safer cardiovascular profile compared to conventional NSAIDs 96 .
Due to the greater widening of the hydrophobic channel of COX-2 to COX-1 by the replacement of isoleucine 523 by another less voluminous vanillin, more voluminous ligands appear to be more selective in inhibiting the catalytic action by the enzyme, since they block more the active site of the enzyme. However, despite the similar volume, eugenol could will be an important lead for the development of new selective COX-2 drugs.
Evaluation of the docking of dual COX1/2-5LOX inhibition with eugenol and NSAIDs (diclofenac and aspirin). In the active site of 5-LOX there are three histidines coordinated to the nonheme catalytic ion Fe 3+ (His 367, His 372 and His 550), plus an isoleucine (Ila 673) and a valine (Val 671). In addition to the iron attachment pocket, a substrate attachment slot contains several hydrophobic residues and additional amino acids that of LOX inhibitors into the pocket. Arachdonic Acid interact in 5-LOX active pocket with the fragments Arg 246, Val 361, Ala 453, Val 243, Leu 244.
Docking studies revealed that the eugenol interacts in the active site of COX-2 as well as with 5-LOX. Eugenol had shown better interaction stabilized of with the active site amino acids of 5-LOX targets respectively the fragments coordinated with Fe 3+ His 367 by van der Waals interactions, His 372 by π-π T-shaped, and Ile 673 www.nature.com/scientificreports/ by H-bonds besides other fragments close to the group coordinated to iron ion as Thr 364 by van der Waals interactions ( Fig. 4 and Supplement 5). In the additional pocket of 5-LOX eugenol interact whit Val 243, Ala 453 by hydrophobic interactions (alkyl and π-alkyl) and by hydrogen bond and electrostatic interaction between eugenol and Arg 370 (Supplement 6). In π-π T-shaped interaction π-electron cloud between the aromatic groups of amino acid fragments and non-chlorinated aromatic ring on diclofenac in a T-shaped manner, i.e., sidewise electron cloud of the ring and head on electron cloud of other ring, these bonds/interactions are necessary to have temporary interactions, especially for the drug action to be accomplished in a system, in addition interactions involving aromatic rings are major contributors to protein-ligand recognition and concomitantly to drug design. Dual COX-2/5-LOX inhibitors had demonstrated excellent analgesic and anti-inflammatory activities with lower gastric irritation, bleeding and ulcerogenic side effects, and are an interesting alternative to provide safer NSAIDs 30,37,43,[97][98][99] . The development of drugs with dual inhibitory activity for COX-2/5-LOX enzymatic pathways offers new options for the development of more effective anti-inflammatory agents with an improved safety profile.
Study in vitro and in vivo models on the cyclooxygenase and lipoxygenase pathways to confirm the results presented, are still necessary. However, computational studies can optimize understanding of how a molecule can act on signaling pathways. Therefore, the computational results can help in elucidating the signal transduction pathways that explain the antiseptic, anesthetic and anti-inflammatory properties of eugenol. The possibility of eugenol acting as an dual inhibitory activity for COX-2/5-LOX enzymatic pathways may shed light on new drugs with structures similar to eugenol that can act in chronic inflammation and control of diseases such as cancer, arthritis, autoimmune, cardiovascular and neurological diseases.

conclusion
The results of molecular prediction for absorption, distribution, metabolism, excretion and toxicity show that eugenol can be used as a pharmacological compound that has similarities to diclofenac and aspirin. Molecular coupling revealed mechanisms of interaction of the eugenol molecule with amino acids at the active sites of COX-2 and 5-LOX. Molecular modeling suggests which regions of amino acids eugenol can act in inhibiting both COX-2 and 5-LOX, being a relevant way to understand the mechanism of inhibition of enzyme activity, indicating the mechanism that eugenol must act physiologically. The possibility of molecular fitting of eugenol with dual activity of COX-2/5-LOX, demonstrating its potential as an anti-inflammatory agent to act in the composition or synthesis of new selective drugs to fight diseases that need to inhibit inflammatory processes such as: osteoarthritis, Crohn's disease and cancer.