Phenolic Compound in Garlic ( Allium sativum ) and Black Garlic Potency as Antigout Using Molecular Docking Approach H )

Article history: Phenolics, including flavonoids, are bioactive components in garlic in relatively abundant amounts and are present 2-4 times more in black garlic. Both of these compounds are reported to have biological activity, with one of them acting as an antioxidant. However , its ability as an antigout is still not widely reported. Xanthine oxidase, adenine deaminase, guanine deaminase, purine nucleoside phosphorylase, and 5-Nucleotidase II are involved during the production of uric acid and causes gout. This study predicted the potential of the phenolic and flavonoid compounds in garlic and black garlic as antigout in inhibiting five target receptors through a molecular docking approach. Utilizing AutoDock Tools v.1.5.7 for receptor and ligand preparation, AutoDock Vina and AutoDock4 for molecular docking, and LigPlot and PyMOL for visualization. About 21compounds from the phenolic and flavonoid groups were used as test ligands and 16 reference ligands (substrate and commercial). SwissADME predicted the pharmacokinetic parameters. The results showed that apigenin, morin, resveratrol, kaempferol, ( + )-catechin, isorhamnetin, and (-)-epicatechin were predicted to have good interactions at each target receptor and had the potential to be developed as candidates for multi-target antigout. Based on the pharmacokinetic parameters, all these compounds had good scores in each, making them feasible to continue in vitro or in vivo trials. Received: 20th May 2022 Revised:18th July 2022 Accepted: 27 July 2022 Online: 31 August 2022


Introduction
Garlic ( Allium sativum L.) is a bulbous flowering plant widely used in culinary and cultivated in various countries in Asia, such as China, Japan, and Indonesia. Garlic bulbs are also reported to have antibacterial, antimicrobial, antiinflammatory, and immunomodulatory properties due to their high sulfur compound content [1,2], In addition to sulfur compounds, phenolic compounds such as flavonoids are reported as natural antioxidants [ 3 ], suitable for treating heart disease and cancer, reducing the risk of chronic diseases, and preventing and treating atherosclerosis [ 4 ], are also available in relatively abundant quantities [ 5 ].
Black garlic is a garlic product that has been heated to high temperatures with controlled humidity for several days. In recent years, black garlic has emerged as a new product with different characteristics from garlic. Black garlic was first introduced in Japan and is gaining popularity in several countries such as China, South Korea, the US, and Europe because of its nutritional content and bioactive components better than garlic [6]. Several sensory properties change during the garlictoblack garlic process, including a darker color due to an increase in melanoidin, a less intense taste and smell due to a decrease in sulfur compounds, and an increase in phenolic, flavonoids, and Sallyl cysteine (SAC) [ 7 , 8], Black garlic is reported to be good for preventing cardiovascular, cancer, obesity, and inflammation [ 6, 9 , 10]. Black garlic has much higher antioxidant activity than garlic because it contains more phenolics, including flavonoids [10]. However, the ability of phenolic compounds in garlic and black garlic as antigout has not been widely reported. Only quercetin in garlic has been reported to reduce blood uric acid levels [11]. Therefore, it is necessary to explore the ability of phenolic compounds, particularly flavonoids, in garlic and black garlic as antigout.
reference ligands consisting of 11 substrates and 5 commercial ligands (Table1and 2). In addition, literature on the ligand was obtained from literature related to garlic and black garlic compounds.
The database used for molecular docking was taken from the Pubchem database catalog (https: // pubchem.ncbi.nlm.nih.gov / ).
The ligand molecule was geometrically optimized using Orca2 software to resemble its natural state. The ligand was prepared using AutodockTools v.1.5. 7 by first adding a hydrogen atom, detecting the root, and selecting the torque. The target protein molecules were taken from the Protein Data Bank (https: // www.rcsb.org / ). The receptors and their natural ligands were separated and converted into PDB format using AutoDock Tools v. 1.5. 7 . Water molecules and other heteroatom molecules were removed. Furthermore, adding polar hydrogen and a Kollman charge after checking for any missing atoms and then saving the file as *.pdbqt. Gout is a condition in which the body synthesizes excessive amounts of uric acid. The synthesis of uric acid in our body is catalyzed by several enzymes, including xanthine oxidase (XDH) [12], adenine deaminase (ADA), guanine deaminase (GDA) [ 13 ], purine nucleoside phosphorylase (PNP) [ 14 ], and 5 -nucleotidase II ( NT 5 C2) [ 15 ]. Gout patients are treated clinically by taking allopurinol and febuxostat, which inhibit XDH activity and thus reduce uric acid production. However, both drugs have side effects such as liver and kidney toxicity [16]. As a result, gout patients frequently have comorbidities such as hypertension, diabetes, and hyperlipidemia [ 17 ].
The decrease in uric acid levels in the blood thus far is achieved by inhibiting a single enzyme, XDH [11,18,19 ]. However, single enzyme inhibition in lowering uric acid levels may be less effective. This study tried to inhibit all enzymes that play a role in uric acid biosynthesis using a molecular docking approach. This method is used because it can predict the binding conformation of ligands to the appropriate target binding site, which is essential in drug design and elucidating fundamental biochemical processes [20]. Therefore, this study aimed to predict the potential of compounds from the phenolic group, including flavonoids in garlic and black garlic as antigout, by inhibiting all target receptors using a molecular docking approach.

Docking molecular simulation
Molecular docking of the receptor was conducted on the test and reference ligands. First, docking was done using AutoDock Vina. The receptor was redocking with cocrystal ligands before docking the test and reference ligands. This redocking aimed to obtain the coordinates of the receptor ' s active site. The structure of the protein and cocrystal ligands was made in the size of a grid box (search space) which covered the entire area of the receptor with sizes x, y, and z, as shown in Table 2 with 1.0 spacing. Grid box size data was stored in the *.config file format.
Then, cocrystal ligand binding to the receptor was done using AutoDock Vina based on the data in the config file. After the redocking was completed, docking validation was performed between the cocrystal and docked ligand, where the validation parameter was the RMSD value < 2 [ 30 ]. After being validated, the test and reference ligands were docked. The docking of the test and reference ligands used a binding site reference area based on the results of redocking. Docking results with AutoDock Vina were analyzed to obtain the highest binding energy. Five ligands with the highest binding energies were docked again using AutoDock 4 . The docking involved creating grid box sizes x = 60, y = 60, and z = 60 with 0.5 spacing. Grid box size data was stored in the *.gpf file format. The selected docking parameters were Genetic Algorithms (GA) run determined as 200 with the population size of 300 , and the maximum number of evals was selected long ( 25 million). GA parameter data was stored in a *.dpf file format. Analysis of docking results in the form of Gibbs free energy and amino acid residue interactions were visualized using LigPlot + 2.2 and PyMOL.

. Screening of bioactive compounds as drug candidates
All compounds targeted by drug candidates were subjected to Lipinski ' s rule of five testing and ADMET (absorption, distribution, metabolism , excretion, and toxicity) testing through the ADMETsar website (http: // www.swissadme.ch / ).

. Molecular docking simulation
Before docking, the docking method must be validated using AutoDock Vina software by redocking the cocrystal ligand to each receptor (Figure1). However, as the xanthine oxidase receptor has no cocrystal ligand , a grid box was created by employing all of the amino acid active sites in the xanthine oxidase protein. The grid used in molecular docking can be seen in Table 3 -The validation parameter was based on the RMSD value, which indicated the level of deviation of the docking ligand position against the cocrystal ligand. The smaller the RMSD value, the smaller the deviation between docking ligand positions and cocrystal ligands, and this position is considered to be the best. The RMSD value is said to be valid if 2 [ 30 ], and if it is in the range of 2 -3 , the validation results are still acceptable [ 31 ].
The redocking of each receptor had an RMSD value of < 2 . This value is acceptable, and the docking process is valid, so the grid box size and position for each receptor can be used for docking with the test ligand [ 32 ]. However, the xanthine oxidase protein does not have an RMSD value because the protein does not have a cocrystal ligand, so the docking method cannot be validated.  (Table 1 and 2). The XDH , ADA, GDA, PNP, and NT 5 C2 were used as target receptors because these five enzymes were directly involved in uric acid biosynthesis [ 24 ]. Figure 2 shows the mechanism of uric acid formation catalyzed by these five enzymes. The Gibbs free energy, or the binding affinity of the ligand to the acceptor, is considered a determining factor in the stability of a ligandprotein complex [ 33 ]. The more negative the Gibbs free energy value or the binding affinity of the ligand to the receptor, the stronger the binding of the ligand to the target receptor, which causes the ligand to be better at inhibiting the target receptor [ 33 ] -The results of the docking simulation using the virtual screening method resulted in different energy affinities at each target receptor are tabulated in Table 4 .
Based on the molecular docking results (Table 4 ), the energy affinity values of the test ligands ranged from -9.30 to -2.70 kcal / mol. The highest energy affinity values of -9.30 kcal / mol belonged to (-)-epicatechin and luteolin compounds when docked to the xanthine oxidase (XDH) receptor. Apigenin, luteolin , myricetin, quercetin, and isorhamnetin are compounds that bind strongly to all target receptors. The value of its energy affinity is more negative than commercial and substrates at each target receptor. The same thing happened to (-)-epicatechin, ( + )-catechin , kaempferol, and morin which were bound to five different receptors, as shown in Table 3 .
Furthermore, caffeic acid binds to XDH, GDA, and NT 5 C2 -1497 receptors, whereas chlorogenic acid binds to XDH, ADA, and GDA receptors. Ferulic acid, pcoumaric acid, and resveratrol are attached to the same two target receptors, XDH and GDA. Finally, mcoumaric acid, ocoumaric acid, and vanillic acid each bind to a single receptor, with mcoumaric acid and ocoumaric acid binding to the XDH receptor. In contrast, vanillic acid binds to the GDA receptor.
Based on molecular docking results of the 21 tested ligands, only phydroxybenzoic acid and gallic acid did not exhibit excellent binding affinities for all target receptors. In comparison, 19 other ligands are suspected as potential candidates for multitarget antigout because 19 compounds had good binding when compared to commercial and substrates at several different receptors based on the binding affinity value. The 19 compounds are highlighted in red (Table 4 ). This study ' s results differ from those previously reported, in which the test ligand compounds from the flavonoid group in garlic were said to only inhibit one target, specifically XDH [11].
A total of 19 test ligands from the docking were ranked, and 5 test ligands were taken from each receptor with the most negative energy affinity value compared to the reference ligands (substrate and commercial). The 5 test ligands with the most negative energy affinity values are highlighted in green ( Table 4 ). The test ligands were docked again using the AutoDock 4 method. The results of the docking are shown in Table  Based on Table 5 , it can be seen that there is a difference in the energy affinity value of the docking results from AutoDock Vina and AutoDock 4 ; this is due to the scoring function between the two methods. The scoring function in AutoDock 4 is semiempirical, involving Coulomb potential, Lennard -Jones potential, system desolvation, and conformational entropy. In contrast, the AutoDock Vina scoring function is empirical, consisting of Gaussian steric, hydrogen, hydrophobic, and covalent bonds [ 34 ]. AutoDock vina has the advantage of docking more quickly and accurately predicting the binding pose than AutoDock 4 . Nonetheless, AutoDock 4 provides good accuracy and precision regarding energy affinity values correlated with experiments [ 34 ].
On the other hand, carcinogenicity indicates the potential of a compound to cause cancer. Positive and negative signs indicate whether or not they can occur [ 38 ]. The last indicator is LD 50 which refers to the maximum dose in milligrams per kilogram of test animal weight that can cause death in test animals. Table 6 shows that allopurinol, azepinomycin , EHNA, fludarabine, and ulodesine as control compounds had good ADMET indicators. Based on the ADMET parameters (Table 6) almost all test compounds had good scores in each parameter, except for the Chlorogenic acid and (-)epigallocatechin gallate had a low score on the bioavailability parameter ( < 0.55 ). Myricetin, luteolin , and quercetin had poor scores on carcinogenicity and AMES mutagenesis parameters, so the five compounds are unsuitable to be recommended as antigout candidates. Finally, seven compounds were found that were suspected to be the best antigout candidates because they had good Lipinski and ADMET parameters. The selection of seven compounds as test ligands is schematically illustrated in the virtual screening protocol, as shown in

. Visual analysis of receptorligand complex interactions
Hydrophobic interactions occurred in the amino acid residues Arg88i and Phe 9 i 5 on the XDH -Allopurinol and XDH -Apigenin complex ( (Figure 4 H). The protein catalytic active site residues are involved in the binding of purine substrates in the mechanism of uric acid formation [ 39 ]. The position of the interacting ligand on the active catalytic site of the receptor causes the receptor activity to be inhibited so that the receptor does not bind to the substrate and cannot form a product (uric acid) [ 24 ]. The interaction of the ligandreceptor complex can be seen in Table 7 .
The visual analysis aims to observe the interactions between the ligands (several test ligands were taken as samples and commercial ligands) as inhibitors and the amino acids of the target protein. Based on the 2D and 3 D visualization results (Figures 4 and 5 ), interactions occurred at the catalytic active site residues in the form of hydrogen bond interactions on the amino acid residue.

. Conclusion
According to molecular docking results, several compounds from the phenolic and flavonoid groups in garlic and black garlic are potential candidates for multitarget antigout therapy. Apigenin and isorhamnetin are the best compounds because they have significantly higher negative energy affinity values in all target receptors (XDH, GDA, PNP, NT 5 C2 -1497 , and NT 5 C2 -1498 ) and have good scores in all Lipinski and ADMET parameters. Other compounds that have been proposed include (-)-epicatechin, ( + )-catechin, kaempferol, and morin, which can inhibit five target receptors and have high scores in all Lipinski and ADMET parameters. Based on the visualization, the (+ )-catechin and morin compounds bind exactly to the active catalytic site to inhibit the product ' s formation in the form of uric acid.
These compounds should be clinically tested against inhibitors at each target receptor. The text in bold is the residue of the active catalytic site -