Chemical composition of leaf essential oil of Schinopsis lorentzii and its inhibitory effects against key enzymes relevant to type-2 diabetes: an emphasis on GC-MS chemical profiling and molecular docking studies

Abstract Schinopsis lorentzii (Griseb.) Engler is an aromatic plant belonging to Anacardiaceae family, and grown in Argentina, Paraguay, Brazil, and Bolivia. It is a rich source of essential oils and tannins as well as commercially available as vegetable-processed leather tanning and antiparasitic agent. This study was designed to explore the chemical composition of essential oil isolated from S. lorentzii leaves and elaborate on their significance as natural anti-diabetic, combined with molecular-docking studies. The leaf oil was chemically analysed using gas chromatography-mass spectrometry (GC-MS) method and investigated for inhibitory effects against α-amylase and α-glucosidase. Molecular-modelling study via Molecular Operating Environment-Docking (MOE-Dock) program was used to evaluate binding interactions of major components with the above-mentioned targets. The GC-MS analysis of the leaf oil revealed the predominance of β-caryophyllene (21.63±0.36%), γ-terpinene (13.53±0.74%), α-pinene (13.38±0.65%), and terpinen-4-ol (12.55±1.56%). The tested oil expressed a remarkable inhibitory effect against α-amylase with IC50 value of 2.27±0.19 μg/mL, comparable to acarbose (IC50 = 0.52±0.18 μg/mL) used as standard drug. Similarly, it significantly inhibited α-glucosidase (IC50 = 0.84±0.03 μg/mL), compared to acarbose (IC50 = 0.38±0.02μg/mL). The in-silico molecular docking study revealed favorable binding affinities of β-caryophyllene, terpinen-4-ol and sabinene with variable degrees with α-amylase and α-glucosidase. Thus, this study demonstrated valuable scientific data on the leaf oil of S. lorentzii as a potential candidate for the development of natural antidiabetic formulations.


Introduction
Diabetic mellitus (DM) is considered as a chronic metabolic disease that is increasing worldwide in prevalence and incidence at an alarming rate 1,2 .The most common type of DM is type-2 diabetes (non-insulin-dependent), which mostly affects adults and reports for 90% of all diabetic patients 3 .Various oral commercial antidiabetic drugs like biguanides, meglitinide, sulfonylureas, thiazolidinedione, dipeptidyl peptidase-4 (DPP-4) inhibitors, sodium-glucose cotransporter, and carbohydrate hydrolyzing enzyme inhibitors have been implemented to control postprandial hyperglycaemia 4,5 .One of the most effective antidiabetic drugs are α-glucosidase and α-amylase inhibitors used for reducing postprandial hyperglycaemia 6 .The available marketable inhibitors such as acarbose, voglibose, and miglitol competitively impede these metabolic enzymes, and thus delay digestion, leading to reduction of carbohydrates' absorption, and this constrain postprandial hyperglycaemia 6 .These drugs are administrated prior to complex carbohydratesrich meals, reduce the glycated hemoglobin (Hb A1c ) levels but these drugs suffer from frequent GIT side effects 7 .Accordingly, several scientists are devoted for discovery of novel α-glucosidase and/or α-amylase inhibitors with fewer or absent undesired effects from natural sources 8 .Nowadays, plant-based medicines and functional foods provide a renewed interest for the prevention and cure of diabetes given their slight or absent side effects 9 .The plant kingdom offers a wide wealthy field of natural bioactive hypoglycemic agents 10 .In the past few decades, over 1200 plant species have been empirically used as hypoglycemic agents worldwide 11,12 .Consequently, the natural inhibitors of α-glucosidase and α-amylase from plant sources are considered as an attractive strategy for treating hyperglycemia.
There are several aromatic plants that produce essential oils, which are complex mixtures of volatile compounds and are employed in different industrial sectors like pharmaceuticals, food, and cosmetics 13,14 .Among essential oils, perfumes and food scents make up the majority of their use 15 .Further, they were found to exhibit different biological effects like antioxidants, antidiabetic, anticancer, antimicrobial, and neuroprotective [16][17][18] .Anacardiaceae embraces about seventy genera and seven hundred species distributed in the tropical and sub-tropical regions 19,20 .The plants of this family are rich in essential oils obtained from a variety of organs, including fruits, leaves, and bark, and exhibit a diverse range of biological effects 21 .The essential oils isolated from Anacardiaceae plants such as Mangifera indica, Mauria heterophylla, Schinus molle, and Pistacia lentiscus are characterized by wealthy existence of β-caryophyllene, α-pinene, γ-terpinene, caryophyllene oxide, terpinen-4ol, myrcene, β-phellandrene, α-humulene and humulene epoxide [22][23][24] .As one of Anacardiaceae sesquiterpenoids, β-caryophyllene was found to control carbohydrate metabolism and improve glycemia control in streptozotocin-induced diabetic rats 25,26 .Schinopsis lorentzii (Griseb.)Engler is a tree belonging to this family spreading in Argentina, Paraguay, Brazil, and Bolivia 27 .The leaves are found to have high content of essential oils, proanthocyanidins, and low content of water-soluble sugars.The European Union (EU-Community Register of Feed Additives) has approved tannin extract as a feed additive 27 .Moreover, the tannin extracts are used to obtain all types of leather and especially natural vegetable-processed leather and also adhesive manufacturing 28 .Further, a commercially available Polyphenolic Vegetable Extract (PVE) from Schinopsis lesions namely "Bioquina ® " was reported for reduction of coccidiosis in broiler chicks 29 .Interestingly, there are no reports regarding the chemical composition of S. lorentzii leaf oil.
This research was directed to analyze the phytocomponents of the essential oil isolated from S. lorentzii leaves via GC-MS technique and investigate α-amylase and α-glucosidase inhibitory properties.Moreover, molecular docking experiments were driven to assess the binding affinities of predominant oil components with the targeted enzymes.

Plant collection and isolation of the leaf essential oil
Fresh leaves of S. lorentzii were purchased from Zohyra Garden in Cairo, Egypt, N 30°02'49.218"E 31°13'32.898".The leaves were gracefully verified by Mrs. Treize Labib, a taxonomy expert at El-Orman Botanical Garden, Giza-Egypt.The leaves (200 gm) were finely cut and hydrodistilled in (2 L water; rotary flask) using Clevenger apparatus for 4 hr and repeated three times.After hydrodistillation, the leaf essential oil was obtained and kept in sealed glass tube at -8°C pending for GC-MS analysis and biological study.

Gas chromatography-mass spectrometry (GC-MS) analysis
The analysis of the leaf essential oil was applied using a TRACE GC-Ultra Gas Chromatography (THERMO Scientific Corp., USA), conjugated with a Thermo mass spectrometer detector (ISQ Single Quadrupole Mass Spectrometer) at the pharmacognosy department, Ain Shams University, Cairo, Egypt.The GC-MS system was equipped with DB-5 column (30 m x 0.25 mm i.d., 0.25 μm film thickness) and helium carrier gas with a flow rate of 1.0 mL/min and a split ratio of 1 to 10.The temperature of oven was programmed at 80°C for 2 min; rising at 5.0°C/ min till 300°C and held for 5 min.The injector and detector were kept at 280°C.About 0.2 μL of diluted oil samples (1:10 hexane, v/v) were injected.The mass spectra were obtained by electron ionization (EI) at 70 eV, using a spectral range of m/z 35-500.

Identification of oil components
The components of the leaf essential oil were identified by matching their mass spectra fragmentation patterns, mass numbers and Kovats retention indices (RI) with those reported in Wily, NIST libraries and the literature data [30][31][32][33][34] .

Assessment of in vitro antidiabetic activities Evaluation of α-amylase inhibition
This investigation was performed using α-amylase Inhibitor Screening Kit (Catalog # K482-100), obtained from Bio-Vision company.Human α-amylase standard was diluted by addition of α-amylase assay buffer (490 μL) into α-amylase enzyme (10 μL) and mixed completely.Then, the diluted α-amylase solution (50 μL) was added to each well containing tested sample, inhibitor control, enzyme control and solvent control.The plate was shaken gently and kept at 25°C for ten minutes, away from light.The α-amylase activity was defined by assessing the mixture absorbance at 405 nm in kinetic mode for twenty-five minutes at 25°C.The inhibition percentage was calculated using the following formula: % Relative inhibition = [(OD control 405 nm -OD sample 405 nm ) / OD control 405 nm ] x 100 where OD is the optical density.Then, the concentration inhibiting 50% of the enzyme (IC 50 ) was calculated from the graph plots of the dose-response curve for each concentration using GraphPad Prism software (San Diego, CA, USA).

Evaluation of α-glucosidase inhibition
This assay was performed using α-glucosidase Inhibitor Screening Kit (Catalog # K938-100), brought from Bio-Vision company.A twentyfold dilution of α-glucosidase was prepared, mixed completely and kept on ice.Diluted α-glucosidase enzyme solution (10 μL) was added to each well containing test samples, inhibitor control, enzyme control and solvent control.The volume of each well reached to 80 μL/well with α-glucosidase assay buffer, then mixed totally and incubated at 25°C for 15-20 min, protected from light.For each well, 20 μL reaction mixture (17 μL buffer + 3 μL substrate) was added to test sample, inhibitor control, enzyme control, solvent control and background control wells and mixed thoroughly.The activity of α-glucosidase was established by assessing the mixture absorbance at 410 nm in kinetic mode for 60 min at 25°C.The inhibition percentage was calculated using the following formula: % Relative inhibition = [(OD control 410 nm -OD sample 410 nm ) / OD control 410 nm ] x 100 where OD is the optical density.Then, the concentration inhibiting 50% of the enzyme (IC 50 ) was calculated from the graph plots of the dose-response curve for each concentration using GraphPad Prism software (San Diego, CA, USA).

In-silico molecular modelling experiments
A molecular docking investigation was done on the identified seven predominant constituents using Molecular Operating Environment (MOE) 2019.0102platform.The structure of the 7 investigated ligands namely, β-caryophyllene, γ-terpinene, α-pinene, terpinen-4-ol, Dlimonene, 2-carene, and sabinene were drawn and converted to MDL mol file format using ChemDraw software.The structures were protonated, and the energies of the identified molecules were minimized using the energy minimization algorithm [gradient: 0.0001, Force Field: MMFF94X].The X-ray crystal structures of selected targets were retrieved from the Protein Data Bank.Initially, the protein molecule was protonated, then unwanted chains and water molecules were removed.Minimization and optimization using force field were performed and the binding sites were identified as the region occupied by the co-crystallized ligand through site finder tool in MOE.Molecular docking simulation was carried out with the MOE Dock module using the induced fit model.This model aims to model the conformational changes that proteins undergo upon binding, including side chain and minor backbone changes to optimize the association 35,36 .The three-dimensional structure of enzyme was set as 'Receptor'.Ligand conformations were generated and the 'Triangle Matcher' algorithm, which is suitable for well-defined binding sites 37 , was set as the ligand placement method.Ten poses were generated for each ligand.All conformations per ligand were scored by the 'London dG' scoring function.The poses were taken through molecular mechanics (MM) refinement to get ten final poses.The final docking score was evaluated with the GBVI/WSA DG scoring function.The best hits were selected based on the S-score and root-mean-square deviation (RMSD) values.The S-score value measures the affinity of the ligand with the receptor.While RMSD compares conformation of docked against reference-docked conformation.Compounds with higher S-value and lower RMSD value are considered the best hits 38 .

In vitro antidiabetic assay
One of therapeutic strategies for DM management is to retard hyperglycemia postingestion which can be achieved by inhibiting enzymes relevant to carbohydrate digestion like α-amylase and α-glucosidase 39 .Thus, α-amylase and α-glycosidase enzymes inhibition assays were used to test in vitro antidiabetic activities of the leaf essential oil (0.10 to 1000 μg/mL) as shown in Table 2 and Fig. 1.
The tested oil and acarbose showed concentration-dependent inhibition of α-amylase and α-glucosidase enzymes.The leaf oil remarkedly inhibited α-amylase with IC 50 value of 2.27±0.19μg/mL, compared to acarbose (α-amylase and α-glycosidase inhibitor standard; IC 50 = 0.52±0.18μg/mL).Further, it showed significant α-glycosidase inhibitory effects with IC 50 value of 0.84±0.03μg/mL, compared to acarbose (IC 50 = 0.35±0.46μg/mL).In this current study, the tested oil demonstrated a large inhibition of 91.23% and 91.70% for α-amylase and α-glucosidase, respectively suggesting its anti-diabetic potential.By inhibiting these enzymes, it could delay carbohydrate digestion and prolong the overall time for carbohydrate digestion resulting to a reduction in the rate of glucose absorption and consequently blunting the post-prandial blood glucose rises, i.e., making the food lower in the glycemic index 40 .The previously reported data on Schinopsis plant extract authenticated its antioxidant and antidiabetic properties 41 .Some of the compounds such as (E)-βcaryophyllene 42,43 , terpinen-4-ol 44 .In the same manner, γ-terpinene 45,46 (Fig. S2) have already been confirmed to possess antidiabetic activities.In the same manner, γ-terpinene 45,46 , have been recently reported to have remarkable in vitro antidiabetic activities.Furthermore α-pinene (0.039 mL/kg) displayed hypoglycemic activity in diabetic animals after 2 and 24 hours of administration using a probit analysis method 16,47 .

Molecular docking studies
Herbal plants can be an efficient source for the development of novel drugs owing to specificity, lower toxicity, target affinity, and plentiful nature.Number of plants with their chemical compounds including flavonoids, phenols and other constituents were effective medicinally in decreasing glucose level in the blood 9,48 .One of the important therapeutic ways to regulate postprandial hyperglycemia in type 2 diabetes mellitus is by inhibiting the metabolism of dietary carbohydrates 49 .First, dietary carbohydrates break down into monosaccharides by alpha amylase activity in the digestive system.The produced monosaccharide is then converted to glucose by α-glucosidase 50 .Therefore, inhibiting α-amylase and α-glucosidase enzyme activity can reduce carbohydrate metabolism, decreasing glucose levels in the blood 51 .In type 2 diabetes, acarbose is used to inhibit glucosidase and amylase 52,53 .Molecular docking simulation is an important method to elucidate the mode of interaction of ligands with specific targets, such as enzymes 54 .In-silico molecular docking is an emerging field to predict and understand possible modes of interaction between a ligand and a target enzyme.The identified potential compounds after GC-MS analysis were used for molecular docking studies to analyze the most favorable conformation and binding affinity with respective targets.X-ray resolved crystal structures of porcine pancreatic alpha amylase (PPA) complexed with acarbose (PDB: 1OSE) 55 and α-glucosidase complexed with acarbose as well (PDB: 3W37) 56 were obtained from Protein Data Bank to evaluate the anti-diabetic potential of the oil major constituents.The compounds were evaluated within the active sites of selected two target enzymes to find out their potential mode of action and binding affinity within the binding sites of the selected enzymes, if any.The results of the docking are presented in Table 3.
The first investigated target was α-amylase.
The enzyme crystal structure was used to study the ability of the selected ligands to be docked in the active site of alpha amylase.To assess the accuracy of MOE-Dock software, the cocrystallized ligand (acarbose) was removed from the active site and redocked within the binding cavity of α-amylase enzyme.Alignment of the pose retrieved from docking to the X-ray (co-crystallized bioactive conformation) was performed to calculate the RMSD value.RMSD value was found as 1.36 Å (Fig. 2), showing that the MOE-Dock method is reliable for docking of the selected ligands 57 .S-score measures the strength of the interaction between the receptor and ligands.The greater the negative value of the S score, the stronger the predicted binding affinity for the enzyme.The docking of acarbose generated an energy score of -14.43 kcal/mol.Acarbose was found to interact with  three catalytically important residues Asp197, Glu233, and Asp300, which are reported as conserved catalytic fold for several families of glycosyl hydrolases 58 .The molecular docking study on the selected ligands and acarbose (as a reference) revealed that all the ten poses of acarbose showed lower S score relative to the selected ligands, suggesting that acarbose will have the strongest binding affinity to the enzyme.Amongst the docked ligands, the top ten docked poses of β-Caryophyllene and terpinen-4-ol showed the best binding energy score relative to the other constituents of the SLL oil.This suggests that these compounds, which showed more poses with lower S score, might have a slightly greater inhibition of alphaamylase.Molecular docking revealed that both compounds were found to have consensus binding mode with co-crystalized acarbose with α-amylase (Fig. 2).β-Caryophyllene was able to form H-π interaction with Trp59 and terpinene-4-ol formed a HB with Asp300.
α-Glucosidases is a member of the glycoside hydrolases family found in the mammalian small intestine involved mainly in starch metabolism, as well as in hydrolyzing maltooligosaccharides prior to their degradation by alkaline phosphatases 59,60 .So, inhibition of the alpha glucosidase enzyme would have an antidiabetic effect.SLL oil significantly inhibited α-glucosidase (IC 50 = 0.65±0.03µg/mL) in comparison with acarbose (IC 50 = 0.38±0.02µg/ mL).To understand the observed α-glucosidase inhibitory activity of the SLL oil, molecular docking analysis study was established to study the binding modes of the target enzyme and docked major constituents of the oil, the number of hydrogen bonds, binding energies of ligand-α-glucosidase complexes and number of closest amino acid residues surrounding the binding site of the α-glucosidase enzyme.All the derivatives formed a complex with the target enzyme.The RMSD value of the correct pose of acarbose was found to be 1.6.The standard deviation of the formed complex for all ligands was less than 1.37 kcal/mol.From Table 3 and Fig. 3, it can be assumed that the maximal inhibition activity was for sabinene, terpinene-4-ol and D-limonene and it is attributable to its binding energy and the stability of the ligand-α-glucosidase complex.The very high binding energy of acarbose (-5.96 kcal/mol) is mainly due to the number of hydroxyl groups and the number of the established hydrogen bonds between the ligand and α-glucosidase enzyme (Fig. 3).Acarbose forms 10 hydrogen bonds with Asp232, Ala234, Asn237, Asp357, Arg552, Asp568 and His626.Besides, two hydrophobic interactions are also observed with residues Phe601 and Trp329.The binding energy of sabinene was found to be the lowest (-3.70 kcal/mol) followed by terpinene-4-ol (-3.67 kcal/mol) among the energies of the docked derivatives.These ligands revealed fitting within the active site.
Sabinene can only form one H-π interaction with Trp329, while terpinene-4-ol was able to form two HB with Asp 357 and His 626 (Fig. 3).

Conclusion and future prospectives
As conclusion, it is the first report on the inhibitory potential of leaf essential oil isolated from S. lorentzii leaves against key enzymes concerned with diabetes.Likewise, this exploration showed a remarkable association between predominant oil components and α-amylase, α-glucosidase inhibitory activities of S. lorentzii.The insilico molecular modeling study exhibited favorable binding affinities of predominant oil components including β-caryophyllene, terpinen-4-ol and sabinene with amylase and glucosidase with varying degrees of affinities.This speculated that these phytocomponents may have contributed significantly to enzyme inhibition activities.Knowing that it is difficult to categorize a single compound responsible for the whole inhibitory activity in these enzymes, we can predict, based on the experimental and in-silico results, that alpha amylase and alpha glucosidase inhibitory activities of oil probably may be owing to the synergistic outcome of these phytoconstituents suggesting their anti-diabetic potential.However, our study was restricted to in vitro biological investigation of the leaf essential oil; therefore, the bioavailability assessment and toxicological profile have not