MAO inhibiting phytochemicals from the roots of Glycyrrhiza glabra L.

Abstract Glycyrrhizin, a natural compound that is substantially present in Glycyrrhiza glabra L. (Gg) root. Monoamine oxidase B (MAOB) inhibitor is used for the treatment of several important neuropsychological diseases like Parkinson’s disease. Gg is known to possess psychoactive properties which relates to its MAO inhibitory potential. This study sought to determine the MAO inhibition property of glycyrrhizin from Gg root extract. The Aqueous extract containing glycyrrhizin was isolated from the root of Gg and characterized using TLC, HPLC, and LC-MS techniques. In silico docking was conducted using Extra precision Glide 2018, Schrödinger docking suite. In addition, the pharmacokinetic properties of the compounds were predicted using SwissADME. The binding energies of the glycyrrhizin correlated well with their in vitro MAO inhibitory potential. Glycyrrhizin exhibited potent inhibitory activity towards MAOB whereas, an aqueous extract of Gg root inhibits both A and B forms of MAO enzyme. Further, molecular docking and molecular dynamics simulation showed that liquiritigenin and methoxyglabridin showed higher stability than other inhibitor compounds from the Gg root extract. These observations suggest that the phytochemicals from the Gg root extract have potent MAO inhibition properties, which can be exploited for the treatment of neurodegenerative disorders. Communicated by Ramaswamy H. Sarma HIGHLIGHTS Catecholamine content decrement in the brain may cause neurobehavioural disorders. MAO may degrade catecholamines in the brain. MAO inhibitors enhance the catecholamine levels. Glycyrrhiza glabra L. extract and their phytoconstituents exhibit MAO inhibiting properties. Glycyrrhizin can be used to treat Catecholamine mediated neurodegenerative disorders.


Introduction
Glycyrrhiza glabra L. (Gg) belongs to the fabaceae family and one of the extensively studied plants in the traditional system of medicine (Mohammad Yousuf Harun et al., 2015).The roots and rhizomes of this plant have been employed clinically for many types of ailments due to its wide-range of therapeutic properties.In the Ayurvedic system, their antiinflammatory (Akamatsu et al., 1991;Siracusa et al., 2011), antiulcer (Dhingra et al., 2004;Prajapati & Patel, 2013), expectorant (Kuang et al., 2018;Nitalikar et al., 2010), antimicrobial (Mitscher et al., 1980), and anxiolytic activities (Ambavade et al., 2001) are described.Along with this, recent studies have shown that Gg possesses psychoactive properties which are reflected in its role as memory enhancer (Dhingra et al., 2004), antidepressant (Dhingra & Sharma, 2006), and antistress (Saxena, 2005) activities.In addition, in vitro analysis of Gg possess monoamine oxidase (MAO) inhibition property (Mazzio et al., 2013).
MAO enzymes play a critical role in the brain in degrading neurotransmitters viz.catecholamines (CA) that includes dopamine (DA), norepinephrine (NE), epinephrine (EN), serotonin or 5-hydroxytryptamine (5-HT) and their precursors (tyramine, 2-phenylethylamine).MAO is a flavin adenosine dinucleotide (FAD) cofactor-containing enzyme located on the outer membrane of the mitochondria and present in two isoforms (Haraguchi et al., 2004).Among two MAO subtypes (MAOA and MAOB), the A subtype is found to have a higher affinity towards NE, 5-HT, and EN while the B subtype degrades DA.Thus, arrest of an enzyme activity by MAO inhibitors (MAOI) exhibits psychoactive properties which are found to be useful in the treatment of panic disorder (Buigues & Vallejo, 1987), a typical depression (Stewart et al., 1992) or mixed anxiety disorder and depression, bulimia and post-traumatic stress disorder (Davidson et al., 1987) as well as a borderline personality disorder (Cornelius et al., 1993).It is also useful in the management of the obsessive-compulsive disorder, trichotillomania, dysmorphophobia, and avoidant personality disorder (B€ urgers et al., 2010).Moreover, it can be used in the treatment of Parkinson's disease (PD) by targeting MAOB in particular (therefore affecting dopaminergic neurons), as well as providing an alternative for migraine prophylaxis (Hamel, 2007).Inhibition of both MAOA and MAOB is used in the treatment of clinical depression and anxiety (Blier & De Montigny, 1994).
MAOI are a group of drugs that inhibits MAO activity of oxidative deamination of biogenic amines (Youdim et al., 2006).According to the selectivity of the drugs towards the type of MAO enzyme, inhibitors can be categorized as MAOA inhibitors (MAOAI) and MAOB inhibitors (MAOBI), while the non-selective inhibitors inhibit both the types of MAO (MAOI) (Youdim & Bakhle, 2006).All these types are further subdivided as reversible or irreversible.Moclobemide and safinamide is a reversible inhibitor of MAOA and MAOB (used for depression and anxiety) (Schapira, 2010;Stein et al., 2002), tranylcypromine, phenelzine, isocarboxazid and selegiline irreversibly inhibit MAOB (used for treating PD) (Bar Am et al., 2004;Laban & Saadabadi, 2022).
These MAOI are commonly used for indications such as a typical depression (Stewart et al., 1992), anxiety disorder, and depression refractory to other therapies or intolerant to other antidepressant drugs (Henriot et al., 1994).MAOB inhibitors are used for PD along with L-DOPA supplementation (Finberg & Rabey, 2016;Nagatsu & Sawada, 2009).Despite their therapeutic value, these drugs exhibit undesirable adverse effects among which edema, insomnia, behavioral changes (especially anxiety or hypomania), autonomic deregulations, sexual dysfunction, and hepatotoxicity are the common problems, whereas hypertensive crisis, hypotension, and hyperpyrexia are potentially risky side effects (Lippman & Nash, 1991).Apart from the synthetic drug, some plant drugs are also known to inhibit the MAO, which is used in traditional medicines and practices.These plant compounds apart from acting as MAOI and also act as antioxidants that increase their therapeutic value (Bhattacharjee & Perumal, 2019).
This study was carried out to determine the MAO inhibitory effect of glycyrrhizin (GgY), a saponin that is the major bioactive compound present in Gg root using molecular docking approach and to compare its MAO inhibitory activity with Gg aqueous root extract (GgE).The study also throws light on the type of MAO inhibited by both GgY and GgE.The compounds from Gg can serve as new therapeutic candidates for better management of neurological conditions, particularly the development of economical, multi-targeted acting drugs against neurodegenerative diseases.

Plant material
Certified commercial dried root powder of Gg was purchased from Rishi Ratna Remedies Pvt. Ltd., Pune, Maharashtra, India.The collected plant specimen was brought to the Botanical Survey of India, Coimbatore for authentication and identified as G. glabra L.-Fabaceae (BSI/SRC/5/23/2019/ Tech/160).A voucher plant specimen was deposited in the DRDO-BU CLS repository for future reference.Glycyrrhizic acid ammonium salt from Glycyrrhiza root (Sigma-AldrichV R , USA) was used as a standard reference.

Preparation of extract -maceration and Soxhlet extraction
GgE was prepared using chloroform-water (0.1%) in the ratio of 1:8 by double maceration process (each maceration for 24 h).The obtained aqueous extract was passed through a muslin cloth and the filtrate was boiled for 5 min.This extract was then set aside for 18 h and filtered using filter paper.The obtained extract fraction was then further concentrated (Dhingra & Sharma, 2005).GgE was also extracted as described earlier (Chopra et al., 2013) with few modifications.A volume of 20 g of the powder was placed inside a thimble made from thick filter paper, which was loaded into the main chamber of the soxhlet apparatus.The extraction was sequentially carried out using petroleum ether (60 � C to 80 � C), chloroform, methanol, ethanol, and water as solvents for 6 to 8 h.The mark in the soxhlet apparatus was dried every time before changing the solvent.The resulting extracts were reduced in vacuum/distilled and stored in glass bottles.

Preliminary phytochemical study
Quality control tests and microbiological assays were outsourced for determining the quality of the purchased Gg root powder, as per the Food Safety and Standards Authority of India (Pardeshi, 2019).

Antioxidant assays
The free radical-scavenging capacity of the plant extract was determined using 1,1-diphenyl-2-picrylhydrazyl (DPPH) stable radical bleaching assay (Arazo et al., 2011;Mohammad Yousuf Harun et al., 2015) and 2,2 0 -azino-bis(3-ethylbenzothiazoline-6sulfonic acid) (ABTS �þ ) radical cation decolourization assay (Siracusa et al., 2011).For the DPPH assay, the reaction mixture (3.5 mL of methanol) contained 100 lM DPPH and different concentrations of extracts; an equal volume of the solvent was employed to dissolve the extracts tested (37.5 lL).The absorbance was recorded at 517 nm after 20 min.Similarly, in the ABTS assay, ABTS �þ was produced by the reaction between 7 mM ABTS in water and 2.45 mM potassium persulfate (1:1) and stored in the dark at room temperature for 12-16 h before use.ABTS �þ solution was then diluted with methanol to obtain an absorbance of 0.7 at 734 nm.To 3.995 mL of diluted ABTS �þ solution, 5 lL of plant extract was added.The absorbance at 734 nm was measured after 30 min initial mixing.An appropriate solvent blank was run in each assay.All the measurements were carried out at least three times.The calculation for the percentage of DPPH or ABTS scavenging effect was done using the general equation, radical-scavenging effect (%)/% Inhibition ¼ A0-A1/A0 � 100 where A0 is the absorbance of the control and A1 is the absorbance of the sample.Trolox was used as a standard substance for both DPPH and ABTS assays.

Thin layer chromatography to confirm the presence of GgY
The GgE was first assessed for the presence of GgY (a saponins derivative) as described previously (Mallikharjuna et al., 2007).The separation of saponins was done using chloroform, glacial acetic acid, methanol, and water (64:34:12:8) solvent mixture.
The colour and Rf values of these spots were recorded by exposing the chromatogram to the iodine vapours.

Measurement of GgY using HPLC analysis
Separation and measurement of GgY in different GgE were achieved using a Phenomenex V R (250 � 4.60 mm 5 m) C18 reversed-phase column.Isocratic elution was carried out using a mobile phase consisting of 0.1% phosphoric acid and acetonitrile in 60:40 v/v, at a flow rate of 1.0 mL/min.The injection volume was 20 lL and elution was carried out at ambient temperature (35 � C) with detection wavelength at 254 nm (Xie et al., 2010).

Protein receptor selection
The crystal structures of both humans (h) and rats (r), hMAOA, hMAOB and rMAOA were obtained from the Protein Data Bank (PDB), with accession ID 2Z5X (hMAOA), 6FW0 (hMAOB) and 1O5W (rMAOA).All the protein structures used were refined to remove the attached extra ligand and then used for docking study.Homology modeling (as per section 2.7) was performed for other proteins with no crystal structures in the database.

Ligand selection
A comprehensive list of 81 phytochemicals present in the aqueous fraction of GgE was selected as per the earlier reports (Asl & Hosseinzadeh, 2008;Chin et al., 2007;Fukai et al., 1998).The selected phytochemicals were used as ligands.The ligand structures were collected from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/).

Homology modeling of rMAOB using SWISS-MODEL
The amino acid sequence of rMAOB (UniProt accession number P19643) was retrieved from the UniProt database and used as a target for homology modeling using the SWISS-MODEL server (Ekins et al., 2016).The server performed the sequence alignment using the best likely match of X-ray template proteins in PDB.The generated 3D models for all target sequences were sorted and the best homology model was selected based on Global Model Quality Estimation (GMQE) and QMEAN statistical parameters.The GMQE data gave the quality metric of target template alignment where in higher values (better values) were chosen.The selected modelled protein was also analyzed for its overall stereochemical quality, including backbone torsional angles through the Ramachandran plot (Bordoli & Schwede, 2012).

Comparative hDAT model construction using I-TASSER
The primary sequence of the target hDAT was obtained from the UniProtKB database (https://www.uniprot.org/uniprot/Q01959) with the sequence ID: Q01959 and entry name SC6A3_HUMAN, sequence length 620 aa.The sequence thus obtained was further submitted to I-TASSER (Iterative Threading ASSEmbly Refinement) server domain (https:// zhanglab.dcmb.med.umich.edu/I-TASSER/)for predicting the three-dimensional structure model from amino acid sequences and structure templates using the fold recognition or threading technique (Roy et al., 2012;Yang & Zhang, 2015;Zhang, 2009).The top I-TASSER hDAT homology model was then modified to include two Na þ ions and one Cl À ion in their reported positions using Maestro (Schr€ odinger, LLC, Portland, OR) (Dahal et al., 2014).The protein was further relaxed, optimized for hydroxyl and thiol groups, and further minimized by aligning the model to the X-ray structure of Drosophila dopamine transporter (PDB ID: 4XP4) while other options remained in default condition in the protein preparation module of the Maestro software package (Maestro Schr€ odinger, LLC, Portland, OR).

Molecular docking
Extra precision Glide 2018, Schr€ odinger docking suite was used for docking studies.All the proteins, modelled proteins, and ligands were processed and docked as mentioned in the following sections.

Protein preparation
Prior to docking, Maestro ver.11.9 protein preparation wizard was used for protein preparation application that executes the corrections like assigning bond orders, add missing hydrogen atoms, finding overlaps, and deleting water molecules within 5 Å distance of MAOA and MAOB structures.In addition to this, the wizard RMSD (root mean square deviation) parameter was set to 0.03 Å, and receptor grid preparation was carried out by optimization of optimized potential for liquid simulation (OPLS3) which systematically includes updated torsional parameters of the force field, additionally increasing the accuracy of predicted energies.For minimizing the chances of nonpolar parts, the Van der Waals radius was scaled to a scaling factor of 1.0 Å and the partial charge cut-off was set to � 0.25 Å.Both the proteins were edited by removal of the inhibitors and chain B of MAO dimer along with the inhibitor associated with chain A of each protein.Other parameters are used as default settings of Maestro ver.11.9.

Ligand preparation
All the retrieved ligand structures from PubChem are subjected to ligand preparation by Ligprep wizard application of Maestro 11.9 (Glide) which Performed corrections like the addition of missing hydrogen atoms, 2D to 3D conversion, correction of bond lengths, angles, low energy structure, stereochemistry and ring conformation followed by minimization of OPLS3 force field.

Molecular docking
Receptor grid-based molecular docking assists the ligand to bind achievable conformation.Molecular docking was carried out using the Maestro-Glide module for the receptor-ligand docking.The prepared and optimized ligand was flexibly docked.The grid box (X ¼ 59, Y ¼ 151.14, Z ¼ 11.73) was used to define the active site pocket of the proteins.The information pertaining to which was obtained from previous reports (Dahal et al., 2014;De Colibus et al., 2005).Top Glide Score (G-Score) and H-bond formation were used to rank the ligands based on their relative binding affinities.The GLIDE module was used to visualize and analyze the specific ligand-protein interaction.

Molecular dynamics (MD) simulation
The molecular dynamic simulations were carried out using the Schr€ odinger LLC-Desmond (Release, 2017a).The NPT ensemble (T ¼ 300 K and p ¼ 1 bar) was used to complete the task.The simulation lasted 100 ns, and the relaxation time for each selected stance was 1 ps.The force field characteristics from OPLS3 were used (Harder et al., 2016).At 10 Å from the protein atoms, orthorhombic periodic box boundaries were constructed, with a cut-off radius of 9.0 for Coulomb interactions.The water molecules were described using the TIP3P model (Jorgensen et al., 1983;Neria et al., 1996).Using Desmond System Builder, the salt concentration was given and changed to 0.15 M NaCl (Release, 2017b).The Martyna-Tuckerman-Klein chain coupling system was used to manage the pressure and has a coupling constant of 2.0 ps.The temperature was controlled by the Nos� e-Hoover chain coupling system (Martyna et al., 1992;1994).All of the gathered data were evaluated and recorded using the previously mentioned methods (Alnajjar et al., 2021;Manoharan et al., 2023).

Binding free energy calculation by MM-GBSA analysis
The MM-GBSA technique of the Prime module from the Schr€ odinger suite was used to determine the binding free energy of the docked complexes, and all other parameters were left at their default values (Shekhar et al., 2019).The binding free energy was calculated using the OPLS_2005 force field, VSGB solvent model, and rotamer search techniques.
The equation represents the calculation of the binding free energy (DG bind ) which is equal to the free energy of the complex (G complex ) minus the sum of the free energy of the target protein (G protein ) and the free energy of the ligand (G ligand ).

PC12 cell cultures
PC12 (pheochromocytoma of the rat adrenal medulla) cells were obtained from the National Centre for Cell Science, Pune.: Non-adherent, non-differentiated PC12 cells were kept frozen in liquid nitrogen until passaged for experiments.The cells were grown in DMEM/F12 (3:1, Himedia TM Laboratories Pvt.Ltd, Mumbai, India) medium supplemented with 15% horse serum, 5% fetal bovine serum, 10% bovine calf serum, 100 units/mL penicillin, 100 units/mL of streptomycin and incubated at 37 � C in an atmosphere of 5% carbon dioxide.The medium was replaced several times per week and cells passaged once per week.

Cell viability assay on exposure to GgY
The viability of PC12 cells was evaluated after 24 h exposure to different concentrations of GgE and the GgY by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) method (Deshetty et al., 2020).Cells were subcultured in 96-well plates at a seeding density of 1 x 10 6 cells per well and were allowed to adhere and grow further.After the cells reached the confluence stage, they were kept in a serum-free medium and further treated using either GgE or GgY after 24 h, the cell survival was calculated by MTT assay.In brief, post-incubation with 20 mL of MTT at 37 � C for 4 h, the MTT formazan crystals containing live cells were solubilized in 200 mL of DMSO.Thereafter, the microwell plate reader measured the absorbance of each well with a test wavelength of 570 nm and a reference wavelength of 630 nm.The IC 50 concentration for GgE and GgY were determined for MTT assay and results were expressed as the mean ± SD of three independent experiments.
Further, the cytotoxicity was evaluated by estimating the level of plasma membrane damage using the lactate dehydrogenase (LDH) release assay quantification kit (Agappe-11407002, India) (Chan et al., 2013).The PC12 cells were grown in 24-well plates with a density of 1 � 10 6 cells/well.After 24 h of adherence, the cells were treated with herbal extract/bioactive compound by fixing the concentration based on IC 50 values obtained from the MTT assay.LDH activity was quantified after 24 h of treatment.The cell suspension was centrifuged at 3,000 �g at 4 � C for 5 min and the supernatant was used for the quantification.LDH leakage at the IC 50 values of herbal extracts and bioactive compounds was quantified and expressed as U/L LDH activity.

Effect of GgE and GgY treatment on MAO activity in PC12 cells
For biochemical enzyme assay the confluent PC12 cells already incubated with the GgE, GgY, and selegiline were centrifuged at 1200 �g for 2 min at 4 � C and homogenized in TES buffer (10 mM Tris-HCl, pH 7.5, 0.5 mM EDTA, 0.25 M sucrose) with micro-pestle and syringe quickly over ice for mitochondrial fraction isolation.The homogenate was centrifuged at 700 �g for 10 min.The resulting pellet including nuclei, unbroken cells, and cell debris was discarded and the supernatant was centrifuged at 7000 �g for 10 mins.The 7000 g pellet (mitochondria fraction) was collected for the MAO assay.AmplexV R (Thermo Fisher Scientific, USA) red MAO assay kit was used to study the inhibitory effect of GgE in the PC12 cell line.The assay was carried out as per the user's manual.MAO-containing samples were diluted in 1 X Reaction Buffer and a volume of 100 lL of tissue samples was used for each reaction.In which HRP mixed benzylamine was used as a substrate of MAOB, and p-tyramine as a substrate of both MAOA and MAOB.For the MAOB inhibitor, pargyline was used to confirm the identity of the enzyme responsible for amine oxidase activity.H 2 O 2 was used as a positive control in the reaction.In addition, resorufin was to prepare a standard curve to determine the moles of product produced in the Amplex Red reaction.The fluorescence was measured in a fluorescence microplate reader (BioTek) using excitation in the range of 530-560 nm and emission detection at �590 nm was used.For each point, background correction for fluorescence was derived by subtracting the values derived from the no-amine oxidase control.

Quality assessment of Gg root powder
Quality control tests and microbiological assays on the purchased Gg root powder were performed as per the Food Safety and Standards Authority of India, which fulfilled the standard norms (Table 1).

Total yield
After maceration and Soxhlet extraction of Gg roots, browncoloured concentrate with 29.4% and 17.3% yield (density: 1.025 g/mL and 3.134 g/mL; total solid contents: 0.053 g/mL and 0.117 g/mL) was obtained, respectively.The vacuumdried extract was used for further analyses.

TLC of GgE
The TLC detection showed the presence of saponins (Rf value ¼ 0.7) upon iodide vapours exposure (Figure 1).

HPLC analysis
After the TLC identification, the GgE was taken for high-pressure chromatographic separation.The HPLC chromatogramprovides information on the possible phytochemicals present in the GgE.GgY gave a highly intense peak at the retention time (RT) of 12.6 min.Using the commercial standard as the control, the quantity of GgY was calculated in the different extracts such as maceration (0.9311 mg/mL), water (0.5966 mg/mL), ethanol (1.7372 mg/mL), pet ether (0.3028 mg/mL), methanol (1.253 mg/mL), and chloroform (0.541 mg/mL) (Supplementary Figure S1.1a).

LC-MS analysis
Further, the LC-MS analysis showed the presence of GgY at RT of 12.09 min (Supplementary Figure S1.2a) with the base peak at m/z of 824 (Supplementary Figure S1.3e) and few additional peaks in the extracted samples.The fragmentation peaks of GgY with m/z of 824, 648, and 454 have confirmed its presence.
The additional peaks and corresponding mass spectra were further analyzed to identify other compounds by referring to the MassBank of North America and Europe databank.The findings revealed other compounds apart from GgY that were present in the extract (Table 2).

In silico analysis of Gg phytochemicals with human monoamine transporters using modelled hDAT
The compounds were docked on the homology-modelled protein of human DAT.The model was created using the I-TASSER server (Supplementary Figure S1.4a-b).Supplementary Figure S1.4c shows the processed modelled protein super imposed to the X-ray structure of Drosophila DAT (PDB ID: 4M48).The protein was quite stable as suggested by the RSMD (6.3 ± 3.9 Å) results and predicted normalized 'B-factor' value.

Molecular dynamics simulation
For the simulation process, the docked compounds were chosen based on the docking score and ADME/Tox analysis.These compounds include Glabridin and liquiritigenin for the protein rMAOA, 3-methoxyglabridin, and 4-O-methylglabridin for rMAOB, Glyzaglabrin and Liquiritigenin for hMAOA,

Binding free energy calculation
Using the MM-GBSA approach, the binding free energies of the complexes hMAOA, rMAOA, hMAOB, and rMAOB were determined (Table 4).The binding free energy was À 35.57kcal/mol for hMAOA complex, À 51.66 kcal/mol for hMAOB complex, À 50.39 kcal/mol for rMAOA and À 43.84 kcal/mol for rMAOB complex.It was found that hMAOB shownthe best binding free energy, which was confirmed by molecular dynamics simulations.The DG bind included coulomb, SolvGB, vdW and other forms of interactions.The phytocompounds liquiritigenin and 3-methoxyglabridin exhibit with maximum DG bind interactions (Supplementary Table S1.6).

Antioxidant analysis of GgE and GgY
DPPH and ABTS assays were performed to assess the radical scavenging potential of GgE and GgY.The GgE exhibited potent radical scavenging activity, however, GgY showed no antiradical activity.Amongst the macerated and Soxhlet extraction methods, the macerated extract showed better antioxidant activity and hence, it was used in further studies (Figure 8a-b).

Effect of GgE and GgY pretreatment on PC12 cell viability
Treatment with GgE and GgY did not cause PC12 cell death up to the concentration of 100 mg/mL, and 15 mg/mL respectively.The IC 50 value of GgE and GgY in PC12 cells was found to be 261.05± 0.87 and 31.80 ± 0.37 mg/mL.Based on the obtained IC 50 value, the concentration of GgE and GgY was further fixed for the MAO inhibition assay (Figure 8c-d).
Additionally, similar results were obtained in the LDH leakage assay, where the control PC12 cells released 25.16 ± 4.72 U/L of LDH.GgE and GgY pretreated cells released LDH ranging between 27.78 ± 3.76 to 238.33 ± 4.89 U/L and 28.16 ± 3.86 to 159.88 ± 4.17 U/L respectively (Figure 8e-f).The dose-dependent LDH leakage was increasedin the GgE and GgYexposed cells indicating damage of the cell membrane, possibly due to oxidative injury and cell death.

Effect of GgE on MAO activity in PC12 cell line
The overall MAO activity was found to be reduced in the GgE exposed cells as compared to GgY.The GgE showed the inhibitory property in both MAOA and MAOB, whereas the MAOB inhibition was prominent when treated with selegiline.However, GgY pretreated cells showed a slight MAOA inhibitory effect, but no significant changes were noted in MAOB activities (Figure 9).

Presence of GgY and other phytochemicals in GgE
GgY constitutes the major bioactive phytochemicals of Gg.This saponin can be separated into the aqueous fraction owing to its high solubility in water (Asl & Hosseinzadeh, 2008).In course of this study, it was observed that GgE contained multiple phytochemicals in notable amounts apart from GgY, which was corroborated by TLC and HPLC findings.This prompted the LC-MS/MS analysis in different extracts, where in the presence of naringenin, glabranin, isoliquitrigenin, liquitrigenin, liquitrin apioside, enoxlone, and GgY was confirmed.The distribution of these phytochemicals was already reported for being associated with Glycyrrhiza species (Asl & Hosseinzadeh, 2008;Chin et al., 2007).In addition, an extensive list of other phytochemicals of the saponin class was also mentioned in earlier reports, which were majorly isolated in the aqueous fraction of the root extract (Fukai et al., 1998).A total of 81 such phytochemicals were thus selected from the literature survey for further in-silico docking analysis.

Drug-likeliness and ADME analysis
During the in-silico pharmacokinetic prediction of these seven compounds using the ADME Swiss web tool (Daina et al., 2017;Manoharan et al., 2022), it came to light that glabranin, isoliquiritigenin, and liquiritigenin can pass the   blood-brain barrier (BBB), whereas naringenin, liquiritin apioside, glycyrrhizin, and glycyrrhetinic acid could not cross BBB.However, our findings for glycyrrhizin, glycyrrhetinic acid, and naringenin contradicted previous publications that claimed the same compounds could transverse the BBB (Lawal et al., 2018;Tabuchi et al., 2012).Therefore, further invitro and in-vivo studies are needed for a proper explanation of such discrepancies.

Docking of GgE phytochemicals with MAO enzymes
In-silico studies revealed that many compounds docked with both the human and rat isoforms of MAO.Among these compounds, naringenin, glabranin, isoliquiritigenin, liquiritigenin, liquiritin apioside, GgY, and glycyrrhetinic acid were found in the LC-MS/MS analysis.Naringenin and isoliquiritigenin were found in GgE (LC-MS/MS) that docked with both the enzymes.Naringenin, a flavanone, was noted to be interacting with the Phe 208 residue of the hMAOA protein.Phe 208 is known to function as a molecular gate in MAOA separating the active site pocket and substrate entry point (Son et al., 2008).Whereas in the rMAOA protein, it interacted with Tyr 407 residue that is part of the active site pocket.Surprisingly, Tyr 407 corresponds to Tyr 326 of hMAOB residue and is known to determine the substrate and inhibitor specificities of the two isoenzymes.Interestingly, naringenin also interacts with hMAOB at Tyr 326 and 398 by pi-pi interactions.Similar observations were made by Son et al. (2008), shows the resemblance of rMAOA to hMAOB protein.
Moreover, previous studies draw a parallel line that naringenin has MAOI properties and can inhibit both isotypes of an enzyme (Govindasamy et al., 2021;Olsen et al., 2008).
Similarly, isoliquiritigenin which gave positive docking interactions with both the isotypes of MAO corraborateswith the findings of Pan et al. (2000) where they found it inhibited rMAO in vitro.Isoliquiritigenin was found to be interacting with Phe 208 and Tyr 407 through pi-pi interactions thus, blocking the substrate entry and active site access to MAOA.Likewise, it bonded with Tyr 326 through hydrogen bonding and pi-pi interaction in hMAOB and rMAOB respectively.Isoliquiritigenin is also known to increase cardiac contraction due to the accumulation of cyclic AMP (Knable & Weinberger, 1997), similar to that of CA (Bhattacharjee et al., 2023;Field et al., 1975;Stangherlin et al., 2011).Further, liquiritigenin was docked with hMAOA (À 10.47), hMAOB (À 10.37), and rMAOB (À 6.95) kcal/mol, which supports with the recent report of Jeong et al. (2020) that it can inhibit both the isoenzymes in vitro and in silico studies.Moreover, in silico docking studies revealed that glabranin and liquiritin apioside can bind to rMAOA.Recent studies suggest that glabranin has antiviral activities against dengue (Ismail & Jusoh, 2017;Sanchez et al., 2000), whereas liquiritin apioside possesses antitussive effects (Kamei et al., 2003;Wei et al., 2020).However, no reports were available on MAO inhibitory properties for these compounds.GgY and glycyrrhetinic acid did not dock any of the MAOs.Overall, the in-silico docking studies revealed that a few of the phytochemicals of GgE have potent binding properties in the active site of MAO.Hence, these compounds might also have MAO inhibitory properties.
Moreover, it was observed that, even though both human and rat MAO proteins are highly homologous to each other, surprisingly 56 ligands specifically docked to rMAOA (not docked with hMAOA counterpart) and 26 ligands specifically docked to hMAOB (not docked with rMAOB counterpart).This can be due to the fact that the active site pocket of rMAOA can accommodate slightly larger and bulkier ligand molecules compared to hMAOA protein.On the other hand, rMAOB was observed to have a smaller pocket with respect to its human counterpart.Ma et al. (2004) have made a similar observation in their studies wherein rat and human MAOA has smaller active site pocket sizes.This suggests that ligands may have inter-species variance in their respective pharmacological response, which should be kept into consideration while developing, designing, testing, and trials pharmacophores.

Docking of GgE phytochemicals with hDAT
Being a channel protein, DAT has an external cavity, which is aqueously accessible.Due to which the majority of the phytocompounds of Gg showed in-silico interactions with the modelled hDAT protein, irrespective of their size.Out of 81 ligands, 71 docked positively to modelled hDAT and showed various molecular interactions with its compact central/primary substrate-binding site residues.Glucoliquiritin apioside (À 13.422 kcal/mol), licuraside (À 11.544 kcal/mol), and neolicuroside (À 10.218 kcal/mol) were found to have the highest docking score.Glucoliquiritin apioside, a flavanone with glucose-apiose sugar moiety at the C 0 -4 position is related to liquiritin apioside and liquiritigenin.It's in-silico docking with hDAT revealed that it forms 8 H-bonds (Asp 79, Arg 85, Asp 313, Thr 316, Phe 320, Asp 476, Ser 539, Arg 544) with À 13.422 kcal/mol docking score and 32 possible poses.This results demonstrate that it might be a potential candidate in terms of interaction with hDAT, wherein it was found to interact with Phe 320 (H-bond).This particular amino acid residue is positioned at the centre of the DAT active site, which basically lies in the interface position (outwardly and inwardly facing conformations of DAT that bind or release substrates on opposite sides of the membrane) (Dahal et al., 2014).In addition, Dahal et al. (2014) found that cocaine analogues interact with Phe 320 residue and showed DAT inhibitory properties.Further, this molecule gets stabilized through hydration site interactions.Licuraside, a chalcone of apioside also docked like glucoliquiritin apioside with Phe 320 residue bonded with the ligand through H-bonding, whereas neolicuroside (a chalcone glycoside) showed pi-pi interactions.However, licuraside had 92 different possible poses of interaction with the hDAT protein.
Apart from these, liquiritin apioside (À 8.766 kcal/mol) also docked to the protein and formed pi-pi and H-bond interaction with the Phe 320 amino acid residue.Liquiritin apioside is already known to have antitussive (cough-relieving or preventive drug) properties through its laryngeal chemoreflex attenuation (Wei et al., 2020).
Interestingly, parallel studies also suggest that at times, when antitussives act as an antidepressant, they reportedly inhibit Gprotein coupled inwardly rectifying potassium channels (Honda et al., 2018) in the nucleus accumbens area of the brain.This inhibitory effect augments the extracellular DA levels (Kawaura et al., 2016).Therefore, liquiritin apioside should be analyzed for either G-protein coupled inwardly rectifying potassium channel inhibition or apioside mediated DAT inhibition in the future.Moreover, glabranin (À 7.236 kcal/mol), isoliquiritigenin (À 6.196 kcal/mol), liquiritigenin (À 6.577 kcal/mol), and glycyrrhetinic acid (À 3.924 kcal/mol) also docked to the hDAT active site.The molecular dynamics simulation studies also revealed the stability of the compounds (3-methoxyglabridin, Glyzaglabrin, Liquiritin, and Liquiritigenin).As discussed earlier, in-vitro studies suggest that glycyrrhetinic acid is the most abundant metabolite actively available in the body after the consumption of Gg and can traverse BBB (Tabuchi et al., 2012).Glycyrrhetinic acid is known to possess psychotropic effects (Kawakami et al., 2010), which might be due to DAT modulation.

Antioxidant properties and cytotoxicity of GgE and GgY
Further, antioxidant properties and cytotoxicity were assessed in GgE and GgY before the in-vitro MAOI studies.The radical scavenging activity using DPPH and ABTS revealed that the GgY did not reduce free radicals at higher than physiological concentrations, as earlier reported by Rackov� a et al. (2007).GgE showed a dose-dependent increase in radical scavenging activity, which was higher than other Soxhlet-extracted samples.These findings were supported by the LDH results.Interestingly, the lower concentration of GgE and GgY supported cell viability, which might be due to the neuroprotective effect (Lee et al., 2008).However, the decrease in viability to 80-85% at higher concentration could be a result of excitotoxicity caused in response to DA accumulation owing to MAO inhibition.An increase in DA can produce either ROS or quinones, which can be detrimental to cells.DA gets metabolized to H 2 O 2 and dihydroxyphenylacetic acid by MAO, then the catechol ring of DA undergoes oxidation to form DA quinone and ROS, resulting in superoxide anion formation.This can occur either spontaneously in the presence of transition metals or enzymatic reaction (Berman & Hastings, 1999).Therefore, the good antioxidant results prompted the use of GgE for further studies.

In vitro MAOA and MAOB inhibition activity of GgE and GgY
In the docking results, GgE compounds and other ligands revealed the different levels of interaction with MAO enzyme isotypes in both rats and humans.After in-vitro analysis of MAOA and MAOB inhibition activity of GgE and GgY, it was clear that the bio-active compound GgY does not show any MAOB inhibition in PC12 cells.However, contrary to the docking results, the slight MAO inhibitory activity of GgY might be due to the metabolic conversion of the compound inside the cell.Further, the presence of either of the three (isoliquiritigenin, liquiritigenin, and/or liquiritin apioside) or more compounds present in trace amounts in the commercial compound might be the reason.This conclusion was drawn based on the presence of a peak at RT 10.08 min in LC-MS/MS.Thus, the MAO assay revealed GgE acts as a potent MAO inhibitor as reported earlier (Mazzio et al., 2013).

Conclusion
Gg is a well-studied herbal plant, which houses a plethora of useful phytochemicals.In the current study, the bioactive compounds in GgE and GgY were predicted to have MAOI and DAT modulatory effects as evident in Molecular docking and Molecular dynamics simulation studies.GgE exhibited potent DPPH and ABTS �þ radical scavenging activities.This study also sheds light on the importance of model selection in-silico and in-vitro as evident from the difference in ligand interaction with rMAO and hMAO protein.

Figure 2 .
Figure 2. Results of docking GgE phytochemicals (confirmed by LC-MS/MS) with MAOA of human (a) and rat (b).A comparative 2D representation of docking results showing substrate, inhibitor, and phytochemical interactions with key active site amino acid residues of MAOA.

Figure 3 .
Figure 3. Results of docking GgE phytochemicals (confirmed by LC-MS/MS) with MAOB of human (a) and rat (b).A comparative 2D representation of docking results showing substrate, inhibitor, and phytochemical interactions with key active site amino acid residues of MAOB.

Figure 8 .
Figure 8. Antioxidant and cytotoxicity assay of GgE and GgY.(a) DPPH and (b) ABTS radical scavenging activity.Cell viability of (c) GgE and (d) GgY treated PC-12 cells.LDH leakage assay in (e) GgE and (f) GgY treated PC-12 cells.Values are expressed as mean ± SEM of three independent experiments.

Figure 9 .
Figure 9.Effect of GgE and GgY treatment on MAO activity in PC12 cells.Values are mean ± SEM of three independent experiments.� p < 0.05, significantly different from control.#p < 0.05 compared to selegiline (one-way ANOVA followed by Tukey's multiple comparison test).

Table 1 .
Quality assessment on Gg Root powder.

Table 2 .
Different RT of most common peaks obtained across the sample type during LC-MS and their identification.
a Highest relative abundance, b Relative abundance of peaks >10%.

Table 3 .
List of selected residues for grid generation.
These interactions varied with each protein despite having almost similar homology.The results of docking MAO inhibitors with hMAOA show that moclobemide gives binding energy score of À 8.350 kcal/mol with the H-bond Phe 208 and Tyr 407.They are compared with the results of Gg phytochemical ligands and found similar H-bond interactions with higher binding affinity (Glyzaglabrin, À 11.021 kcal/mol with H-bond Phe 208, Tyr 407, Tyr 444, Gly 443; Liquiritigenin, À 10.477 kcal/mol with H-bond Phe 208; Naringenin, À 10.404 kcal/mol with Hbond Phe 208, Cys 323, Asn 181).Additionally, it is suggested that further in-vitro and in-vivo studies should be performed to reveal the exact reason for the discrepancies in differential MAO inhibitory action of the ligands.Many GgErelated lead compounds can be utilized for pharmacological drug design and development against MAO enzyme or DAT.Also, specific phytochemicals from GgE should be tested with brain region-specific drugs and inhibitors to elucidate the underlying molecular mechanisms in more detail.Such analysis would provide insights into these ligands that might possess selective inhibitory properties and may be helpful in drug design.