Evaluation of Major Constituents of Medicinally Important Plants for Anti-Inflammatory, Antidiabetic and AGEs Inhibiting Properties: In Vitro and Simulatory Evidence

Diabetes mellitus (DM) is a global health concern that is associated with several micro- and macrovascular complications. We evaluated several important medicinal plant constituents, including polyphenols and flavonoids, for α-glucosidase inhibition, AGEs’ inhibitory activities using oxidative and no-oxidative assays, the inhibition of protein cross link formation, 15-lipoxydenase inhibition and molecular docking. The molecular docking studies showed high binding energies of flavonoids for transcriptional regulars 1IK3, 3TOP and 4F5S. In the α-glucosidase inhibition assay, a significant inhibition was noted for quercitrin (IC50 7.6 µg/mL) and gallic acid (IC50 8.2 µg/mL). In the AGEs inhibition assays, quercetin showed significant results in both non-oxidative and (IC50 0.04 mg/mL) and oxidative assays (IC50 0.051 mg/mL). Furthermore, quercitrin showed inhibitory activity in the non-oxidative (IC50 0.05 mg/mL) and oxidative assays (IC50 0.34 mg/mL). A significant inhibition of protein cross link formation was observed by SDS-PAGE analysis. Quercitrin (65%) and quercetin (62%) showed significant inhibition of 15-lipoxygenase. It was thus concluded that flavonoids and other polyphenols present in plant extracts can be effective in management of diabetes and allied co-morbidities.


Introduction
Diabetes mellitus (DM) is an ailment of the endocrine system that is associated with chronic insulin resistance and progressive exhaustion and death of β-cells in the pancreas that leads to hyperglycemia [1]. At present, 382 million people are currently diagnosed with diabetes globally with an expected massive increase to 600 million by the year 2035 [2]. About 4.9 million people die every year as a result of diabetes and 50% of this death toll is a consequence of diabetic complications [3]. In diabetes, several mechanisms initiate and impart injury to different vascular structures and repair mechanisms in such patients [4]. Thus, patients suffering from diabetes are susceptible to intense long-standing impediments like atherosclerosis, impaired wound healing, neuropathy, retinopathy, periodontitis, cataracts and nephropathy [5][6][7].
Advanced Glycation End products (AGE)s are the products produced via a chain of non-enzymatic reactions between reducing sugars, such as glucose, and the amino functionalities of proteins [8]. This reaction, also known as the Maillard reaction, starts when

α-Glucosidase Assay
The extracts and pure compounds were assessed for their potential antidiabetic activities using the α-glucosidase assay. Amongst all tested extracts, only Juglans regia (61%) and the peel of Punica granatum (52%) showed the inhibition of α-glucosidase (Table 4). All other plants were considered inactive, since only mild inhibition was seen. In the case

Antiglycation Assays
In the AGEs inhibition assays, both oxidative (BSA-MGO) and non-oxidative (BSAglucose) modes of inhibition were analyzed (Table 6). Among flavonoids, quercetin presented significant results in both the non-oxidative and (IC 50 0.04 mg/mL) oxidative assay (IC 50 0.051 mg/mL). Quercitrin showed a better inhibition in the non-oxidative (IC 50 0.09 mg/mL) than in the oxidative assay (IC 50 0.34 mg/mL) ( Table 6). Apigenin (IC 50 0.45 mg/mL) was only active in the non-oxidative mode. Juglone has shown its activity in both the oxidative (IC 50 0.11 mg/mL) and non-oxidative (IC 50 0.06 mg/mL) assays, which reflects its potential use for diabetes and its complications ( Table 6). All other tested compounds were recorded as inactive.

Protein Cross-Linking Assasy
The compounds were further tested for the inhibition of cross-link formation using both the non-oxidative (BSA-glucose) and oxidative (BSA-MGO) modes. SDS-PAGE was employed for measuring cross-linking. BSA incubation with either glucose or MGO produces small amounts of dimerization that is not visible with incubation of BSA alone. All tested compounds showed inhibition of cross-linked AGEs causing a decrease in intensity of dimerized (cross-linked) bands ( Figure 4).

Protein Cross-Linking Assasy
The compounds were further tested for the inhibition of cross-link formation using both the non-oxidative (BSA-glucose) and oxidative (BSA-MGO) modes. SDS-PAGE was employed for measuring cross-linking. BSA incubation with either glucose or MGO produces small amounts of dimerization that is not visible with incubation of BSA alone. All tested compounds showed inhibition of cross-linked AGEs causing a decrease in intensity of dimerized (cross-linked) bands ( Figure 4). In the BSA-MGO assay, amongst all tested compounds, 2-phenylethylisothiocyanate presented the highest inhibition (48%) followed by α-humulene (40%), juglone (39%) and caryophyllene oxide (35%). The flavonoids showed a mild inhibition ranging from 21-34% ( Figure 5). It was thus concluded that all tested compounds showed a mild inhibition of protein cross-link formation. In the BSA-MGO assay, amongst all tested compounds, 2-phenylethylisothiocyanate presented the highest inhibition (48%) followed by α-humulene (40%), juglone (39%) and caryophyllene oxide (35%). The flavonoids showed a mild inhibition ranging from 21-34% ( Figure 5). It was thus concluded that all tested compounds showed a mild inhibition of protein cross-link formation. In the BSA-glucose assay (Figure 6), amongst all tested compounds, quercitrin presented the highest inhibition (55%), followed by quercetin (53%), and apigenin (51%). The flavonoids showed a mild inhibition ranging from 36-46% (Figure 7). It was thus concluded that all tested compounds showed a mild inhibition of protein cross-link formation.  In the BSA-glucose assay (Figure 6), amongst all tested compounds, quercitrin presented the highest inhibition (55%), followed by quercetin (53%), and apigenin (51%). The flavonoids showed a mild inhibition ranging from 36-46% (Figure 7). It was thus concluded that all tested compounds showed a mild inhibition of protein cross-link formation. In the BSA-glucose assay (Figure 6), amongst all tested compounds, quercitrin presented the highest inhibition (55%), followed by quercetin (53%), and apigenin (51%). The flavonoids showed a mild inhibition ranging from 36-46% (Figure 7). It was thus concluded that all tested compounds showed a mild inhibition of protein cross-link formation.

Discussion
In this study, the effect of medicinal plant extracts and their major components on diabetes, glycation and inflammation was demonstrated using in vitro and in silico methods. Structural diversification of natural compounds enables multiple biological activities due to diverse mechanisms of action. Polyphenols and flavonoids possess several health benefits on account of their antioxidant, anti-inflammatory, and enzyme inhibiting properties [39]. Generally, the strong antioxidant potential of such compounds may contribute towards antidiabetic, antiglycation and anti-inflammatory activities [40,41].
α-Glucosidases are enzymes located at the intestinal lumen brush border that catalyze the hydrolysis of terminal, non-reducing α-1-4-linked glucose residues of disaccharides or oligosaccharides [42]. These enzymes are therefore helpful to facilitating carbohydrate absorption. Inhibitors of α-glycosidase can interrupt the digestion of carbohydrates to glucose and therefore they can be used for the treatment of type 2 diabetes [43]. In this study, flavonoids and polyphenols exhibited significant inhibition of α-glucosidase, which may be attributed to the presence of OH groups in the C-ring (flavonoids). It has been reported earlier that 3-hydroxylation of the C-ring facilitates the inhibitory activity against α-glucosidase [44]. The data is also in agreement with the molecular docking assessment, where strong H-bonding and hydrophobic interactions were observed for the transcriptional regulator gene for 3TOP (α-glucosidase).
Tissue inflammation is primarily related to immune system activation in diabetic patients [45]. Further phenotype conversion of macrophages from M2-type to M1-type is also very important in inflammatory conditions [46]. Lipoxygenase activation (12, 15lipoxygenase) also plays a key role in the development of diabetes, and evidence has suggested that Lox inhibitors can greatly protect against diabetes [47]. Our results indicated that flavonoids and polyphenolic compounds possess significant inhibition of 15-lipopxygenase. This is further supported by molecular docking results, which indicate strong Hbonding, and hydrophobic interactions with transcription regulator 3TOP. This could be due to pi-interaction of the bond linking the B and C rings [40] that gives a near planar region of these two rings. Such structures (like flavonoids) easily enter the hydrophobic pockets in enzymes and can subsequently increase their inhibitory effect. Thus the C2-C3 double bond of flavonoids is crucial for their anti-inflammatory activity [48].

Discussion
In this study, the effect of medicinal plant extracts and their major components on diabetes, glycation and inflammation was demonstrated using in vitro and in silico methods. Structural diversification of natural compounds enables multiple biological activities due to diverse mechanisms of action. Polyphenols and flavonoids possess several health benefits on account of their antioxidant, anti-inflammatory, and enzyme inhibiting properties [39]. Generally, the strong antioxidant potential of such compounds may contribute towards antidiabetic, antiglycation and anti-inflammatory activities [40,41].
α-Glucosidases are enzymes located at the intestinal lumen brush border that catalyze the hydrolysis of terminal, non-reducing α-1-4-linked glucose residues of disaccharides or oligosaccharides [42]. These enzymes are therefore helpful to facilitating carbohydrate absorption. Inhibitors of α-glycosidase can interrupt the digestion of carbohydrates to glucose and therefore they can be used for the treatment of type 2 diabetes [43]. In this study, flavonoids and polyphenols exhibited significant inhibition of α-glucosidase, which may be attributed to the presence of OH groups in the C-ring (flavonoids). It has been reported earlier that 3-hydroxylation of the C-ring facilitates the inhibitory activity against α-glucosidase [44]. The data is also in agreement with the molecular docking assessment, where strong H-bonding and hydrophobic interactions were observed for the transcriptional regulator gene for 3TOP (α-glucosidase).
Tissue inflammation is primarily related to immune system activation in diabetic patients [45]. Further phenotype conversion of macrophages from M2-type to M1-type is also very important in inflammatory conditions [46]. Lipoxygenase activation (12, 15-lipoxygenase) also plays a key role in the development of diabetes, and evidence has suggested that Lox inhibitors can greatly protect against diabetes [47]. Our results indicated that flavonoids and polyphenolic compounds possess significant inhibition of 15-lipopxygenase. This is further supported by molecular docking results, which indicate strong H-bonding, and hydrophobic interactions with transcription regulator 3TOP. This could be due to pi-interaction of the bond linking the B and C rings [40] that gives a near planar region of these two rings. Such structures (like flavonoids) easily enter the hydrophobic pockets in enzymes and can subsequently increase their inhibitory effect. Thus the C2-C3 double bond of flavonoids is crucial for their anti-inflammatory activity [48].
AGEs build-up in the body can activate several signaling pathways through receptors and thus interrupt various biological activities and cellular functions that finally lead to cell death [49]. Various mechanisms have been documented to explain AGEs synthesis in the human body including oxidative and non-oxidative modes [50]. In this investigation, flavonoids are reported with antiglycation activities in both oxidative and non-oxidative models that may be due to their radical scavenging properties, thus a delay in the progression of glycation is expected [51,52]. The cross-linking AGEs possess strong affinity towards diverse proteins and therefore are resistant to degradation. This leads to further toxicity to the human body [53]. Furthermore, flavonoids also showed inhibition of protein cross-link formation. It was evident from earlier findings that flavonoids and polyphenols inhibit cross-link formation because of antioxidant properties and some other mechanisms [54]. Thus, radical scavenging and inhibition of cross-link formation may provide a protective effect against hyperglycemia-mediated damaging effects on proteins [55]. Our in silico results also supported this fact by indicating strong H-bonding interactions of flavonoids and polyphenols with transcriptional regulator 4F5S (BSA).  (Table S1). The plants were analyzed as reported before using HPLC-DAD, GC-MS, LC-QTOF-MS [56], and major components are highlighted in the Supplementary Materials ( Figure S1).

Extraction and Drying
Plant material was dried in an oven below 40 • C. Next, the plant material was thoroughly grinded followed by cold maceration in 90% methanol. Solvent evaporation was accomplished by a rotary evaporator (Büchi, Flawil, Switzerland), and stored at 4 • C until use. The essential oils were obtained using a Clevenger apparatus (hydrodistillation) and fixed oils were obtained using cold pressing [56].

Molecular Docking
For molecular docking studies, the X-ray crystallographic structures of the transcriptional regulators 3TOP [57] and 1IK3 [58] were taken from the Protein Data Bank (PDB) and active pocket dimensions for each protein were checked using the CASTp 3.0 online tool. The optimization of transcriptional regulators was performed using DS Visualizer 2.0 [59]. Furthermore, the structures of all the phytoconstituents were downloaded from the Pubchem database and PDB files were generated in DS Visualizer 2.0 [59]. The molecular docking was performed using Lamarckian Genetic Algorithm embedded in AutoDock v 4.2 [60]. A total number of nine poses were generated (for each target) and grouped according to their RMSD values. Every set was prudently checked in Discovery Studio Visualizer and presumed binding modes were highlighted for further analysis. Best docked structures based on the binding energy scores (∆G) and H-binding were chosen for further analysis. The hydrogen bonding and hydrophobic interactions between ligand and protein were calculated by Ligplot + and DS Visualizer 2.0.

α-Glucosidase Inhibition Assay
The α-glucosidase inhibition experiment was accomplished by using a modified method [61]. Initially, the enzyme solution (from Saccharomyces cerevisiae) (0.2 units/mL dissolved in 0.1 M phosphate buffer; pH 6.8) was mixed with the test sample (1 to 0.039 mg/mL) and incubated in an oven (37 • C for 10 min). After incubation, the substrate (p-nitrophenyl-α-D-glucopyranoside; 0.29 mM) was added to the enzyme and test sample solution and incubated for another 30 min (37 • C). The reaction was halted by adding Na 2 CO 3 (100 µL, 200 mM stock) to this mixture and absorbance was noted at 400 nm. Acarbose was used as positive control.
The percentage of inhibition was determined using the following formula: % The AGEs assay was accomplished by using a standard protocol [62,63]. The protein source (bovine serum albumin, BSA) (10 mg/mL; 135 µL) was mixed with D-glucose solution (500 mM, 135 µL in phosphate buffer; 50 mM, pH 7.4), NaN 3 (sodium azide; 0.02%) and test samples (various concentrations). The reaction mixtures were kept at 60 • C for 1 week to facilitate glycation. Finally, trichloroacetic acid (10 µL, 100%) was added to it to stop the reaction and precipitate the unbound material. The supernatant was removed and the pellet was dissolved in alkaline phosphate buffer saline (137 mM NaCl, 8.1 mM Na2HPO4, 2.68 mM KCl, 1.47 mM KH2PO4, pH10). Finally the fluorescence intensity (ex 370/emis 440; ex 335/emis 385 nm) was recorded on a spectrofluorometer (FLx800, BioTek, Winooski, USA). Aminoguanidine was used as positive control. The control samples used for the experiment were prepared using the same protocol without the test sample.
The AGEs inhibition was calculated as where Fluo is the fluorescence intensity. The IC 50 was calculated using MS Excel.

BSA-MGO Assay
In this oxidative glycation assay, the protein source (BSA) was mixed (10 mg/mL, 135 µL) with methylglyoxal (5.75 mM, 135 µL) and liquefied in phosphate buffer (50 mM, pH 7.4) and NaN 3 (sodium azide; 0.02%). The test samples were added to this reaction mixture and stored at 37 • C for a week. The change in fluorescence intensity (ex 370/emis 440; ex 335/emis 385 nm) was recorded on a spectrofluorometer (FLUOstar Omega ® , BMG Lab Tech, Aylesbury, UK). Aminoguanidine and quercetin were used as positive controls. The control samples used for the experiment were prepared using the same protocol without a test sample.
The where Fluo is the fluorescence intensity. The IC 50 was calculated using MS Excel.

SDS-PAGE Gels Image Analysis
The developed gel was stained with Coomassie blue and images were obtained using GelDoc. Finally, the Image J tool was used for the determination of integrated density (IntDen). The integrated density was employed further for the determination of percentage inhibition. The integrated density (ID) was determined as follows: where N is the number of pixels in the selection and the background is the modal grey value (most common pixel value) after smoothing the histogram.
The percentage inhibition of cross-linked AGEs was determined using the formula: % inhibition = 100 × (ID without inhibitor − ID with inhibitor)/ID without inhibitor

15-Lipoxygenase Assay
The anti-inflammatory activity of test samples was evaluated using a standard protocol [65]. To the enzyme solution (200 units/mL; 487.5 µL), different concentrations of test sample (12.5 µL (2-0.062 mM) in DMSO were added and incubated for 5 min at 37 • C. The absorbance was recorded immediately after the addition of substrate (500 µL of substrate (250 mM linoleic acid in 0.2 M borate buffer, pH 9) and after every min up to 5 min at 234 nm by using a UV spectrophotometer (UV-1601, SHEMADZU, Kyoto, Japan).
The percentage inhibition of enzyme activity was calculated as follows: where ∆A1/∆t and ∆A2/∆t are the increase rate in absorbance at 234 nm for a sample without test substance and with test substance, respectively.

Statistical Analysis
All experiments were performed in triplicate and results were expressed as mean ± SD.

Conclusions
Diabetes, a very prevalent metabolic disorder, is an important health problem worldwide. Persistent hyperglycemia results in the development of inflammation and several life threatening complications due to the production of AGEs. Various strategies are being used by researchers to introduce new treatment options that have a dual effect i.e., both blood glucose lowering and anti-AGEs potential. Traditional medicinal plants are considered as effective and reliable alternatives of conventional medical therapy due to proven safety and efficacy. In this investigation, selected medical plants and their major constituents were analyzed using in silico and in vitro models. Interactions of flavonoids and polyphenols were observed with transcription regulators 1IK3, 3TOP and 4F5S. Further in vitro assays presented anti diabetic, antiglycation, and anti-inflammatory activities of constituents including juglone, quercetin, quercitrin, apigenin and 2-phenylethylisothiocyanate from Juglans regia, Punica granatum and Myristica fragrans. Thus, it was concluded that these plant species may be considered as candidates for the management of diabetes mellitus and co-morbidities occurring due to AGEs.  Table S1. Ayurvedic medical plants and their major constituents. Figure S1. Structure of test compounds. Figure S2. 3D H-bonding interactions of 2-phenyl isothiocyanate pose no. 1 [1], Apigenin pose no. 5 [2] caryophyllene oxide pose no. 1 [3] with binding sites of transcriptional regulator 1IK3. Figure S3. 3D H-bonding interactions of eugenol pose no. 3 [4], α-humulene pose no. 1 [5] with binding sites of transcriptional regulator 1IK3. Figure S4. 3D H-bonding interactions of quercitrin pose no. 4 [6] and caryophyllene pose no. 1 [7] with binding sites of transcriptional regulator 1IK3. Figure S5. 3D H-bonding Interactions of 2-phenylethylisothiocyanate pose no. 1 [8], apigenin pose no. 1 [9]; caryophyllene oxide pose no. 1 [10] with binding sites of transcriptional regulator 3TOP. Figure S6. 3D H-bonding interactions of eugenol pose no. 2 [11] and α-humulene pose no. 1 [12] with binding sites of transcriptional regulator 3TOP. Figure S6. 3D H-bonding interactions of quercitrin pose no. 4 [13] and caryophyllene pose no. 1 [14] with binding sites of transcriptional regulator 3TOP. Figure S7. 3D H-bonding interactions of 2-phenylethylisothiocyanate pose no. 1 [15], apigenin pose no. 3 [16] and caryophyllene oxide pose no. 1 [17] with binding sites of transcriptional regulator 4F5S. Figure S9. 3D H-bonding interactions of quercitrin pose no. 5 [20] and caryophyllene pose no. 1 [21] with binding sites of transcriptional regulator 4F5S [66][67][68][69].