In Silico Molecular Docking and In Vitro Antidiabetic Studies of Dihydropyrimido[4,5-a]acridin-2-amines

An in vitro antidiabetic activity on α-amylase and α–glucosidase activity of novel 10-chloro-4-(2-chlorophenyl)-12-phenyl-5,6-dihydropyrimido[4,5-a]acridin-2-amines (3a–3f) were evaluated. Structures of the synthesized molecules were studied by FT-IR, 1H NMR, 13C NMR, EI-MS, and single crystal X-ray structural analysis data. An in silico molecular docking was performed on synthesized molecules (3a–3f). Overall studies indicate that compound 3e is a promising compound leading to the development of selective inhibition of α-amylase and α-glucosidase.

Due to numerous biological applications of 2-aminopyrimidine and acridine amine and its analogues, we have focused our research on synthesis of 2-aminopyrimidine by using pharmacologically important structural scaffold acridine pharmacophore.
Protein-ligand interaction is comparable to the lock-andkey principle, in which the lock encodes the protein and the key is grouped with the ligand. The major driving force for binding appears to be hydrophobic interaction [24]. In silico techniques helps identifying drug target via bioinformatics tools. They can also be used to explore the target structures for possible active sites, generate candidate molecules, dock these molecules with the target, rank them according to their binding affinities, and further optimize the molecules to improve binding characteristics [25]. Diabetes mellitus (DM) is a leading noncommunicable disease which affects more than 100 million people worldwide and is considered as one of the fine leading diseases which causes death in the world [26]. Type-2 diabetes mellitus is a chronic metabolic disorder that results from defects in both insulin secretion and insulin action. Management of type-2 diabetes by conventional therapy involves the inhibition of degradation of dietary starch by glucosidases such as -amylase and -glucosidase [27]. The pathogenesis of type-2 diabetes involves progressive development in insulin resistance associated with a defect in insulin secretion, leading to overt hyperglycemia. However, compounds which improve insulin sensitivity and glucose intolerance are somewhat limited warranting the discovery and characterization of novel molecules targeting various pathways involved in the pathogenesis of type-2 diabetes [28]. Currently, there are few drugs that are able to counteract the development of the associated pathologies. Therefore, the need to search for new drug candidates in this field appears to be critical. Aromatic amines and thiazolidinones nuclei would produce new compounds with significant antidiabetic properties [29].

Chemistry.
Melting points were determined by open capillary method and are corrected with standard benzoic acid. All solvents were distilled and dried prior to use. TLC was performed on silica gel G and the spots were exposed to iodine vapour for visualization. A mixture of petroleum ether and ethyl acetate was used as an eluent at different ratio. Column chromatography was performed by using silica gel (60-120 mesh). 1 H NMR and 13 C NMR spectra were recorded in CDCl 3 on a Bruker advance 400 MHz instrument. Chemical shifts are reported in ppm using TMS as the internal standard. IR spectra were obtained on a Perkin-Elmer spectrum RXI FT-IR spectrometer (400-4000 cm −1 ; resolution: 1 cm −1 ) using KBr pellets. Molecular mass was determined using ESI-MS THERMO FLEET spectometer.
Spectral data of the synthesized compounds are described below.

Single Crystal X-Ray Diffraction (XRD) Analysis.
Crystals suitable for X-ray analysis were obtained by slow evaporation of a solution of the 10-chloro-4-(2-chlorophenyl)-12phenyl-5,6-dihydropyrimido[4,5-a]acridin-2-amine, 3f, in ethyl acetate. The measurements were made on Enraf Nonius CAD 4 -MV 31 single crystal X-ray diffractometer. The diagrams and calculations have been performed using SAINT (APEX II) for frame integration, SHELXTL for structure solution and refinement software programs. The structure was refined using the full-matrix least squares procedures on F 2 with anisotropic thermal parameters for all nonhydrogen atoms. The crystal data and details concerning data collection and structure refinement of compound, 3f, single crystal are summarized in Table 2. As we can see compound 3f crystallizes in triclinic space group, P-1. The single crystal structure and atomic numbering chosen for compound 3f are demonstrated in Figure 2. Selected bond lengths and bond angles are compiled in Table 3. It can be observed that the bond lengths of all kind atoms on the whole molecule are absorbed. The typical C-C single (1.55Å) and C=C double (1.38Å) bonds, C-N typical single (1.35Å) and C=N double (1.33Å) bonds. This means that the carbon-carbon bond and the carbon-nitrogen bond have double bond character and contribute to form conjugated system. Furthermore, carbon attached with strong electronegative atom chlorine bond length is C-Cl (1.68 A), in case all C-H (0.97 A) and C=H (0.93 A) are absorbed. The primary amine hydrogen bond length is N-H (0.86 A), respectively (Table 3). Concerning inspection of the torsion angles, the 3f molecule has nearly planar core. Figure 3 illustrates a representative view of compound 3f crystal packing structure. The two molecules of compound 3f are packed in face-to-face arrangement due to the intermolecular hydrogen bonds. The crystal packing structure demonstrates the existence of two intermolecular hydrogen bonds. Interesting result were absorbed from crystal packing structure is the intermolecular hydrogen bonding occurs at C (23) Figure 3: The packing of the molecules in the unit cell, viewed down the a-axis with the hydrogen bond geometry of compound 3f.

Glucose Diffusion Inhibitory Test
Sample Preparation.

Inhibition Assay for -Amylase Activity.
Four different concentrations (100, 200, 300, and 400 g/mL) of samples and standard drug acarbose were prepared and made up to 1 mL with DMSO. A total of 500 L of sample and 500 L of 0.02 M sodium phosphate buffer (pH 6.9 with 0.006 M NaCl) containing -amylase solution (0.5 mg/mL) were incubated for 10 minutes, at 25 ∘ C. After preincubation, 500 L of 1% starch solution in 0.02 M sodium phosphate buffer (pH 6.9 with 0.006 M NaCl) was added to each tube. This reaction mixture was then incubated for 10 minutes at 25 ∘ C. 1 mL of DNSA colour reagent was added to stop the reaction. These test tubes were then incubated in a boiling water bath for 5 minutes and cooled to room temperature. Finally this reaction mixture was again diluted by adding 10 mL of distilled water. % of inhibition by -amylase can be calculated by using the following formula. Absorbance was measured at 540 nm [32,33]: Triplicates were done for each sample at different concentrations.

Inhibition Assay for -Glucosidase Activity.
Various concentrations of samples and standard drug acarbose were prepared. -Glucosidase (0.075 units) was premixed with   sample. 3 mM p-nitrophenyl glucopyranoside used as a substrate was added to the reaction mixture to start the reaction [34]. The reaction was incubated at 37 ∘ C for 30 min and stopped by adding 2 mL of Na 2 CO 3 . The -glucosidase activity was measured by p-nitrophenol release from PNPG at 400 nm. % of inhibition can be calculated by using (1). Triplicates are done for each sample at different concentrations [31][32][33].

Statistical
Analysis. Statistical analysis was performed using one-way analysis of variance (ANOVA). Results are expressed as mean ± SD and = 3.

Glucose Diffusion Inhibitory Test.
The results are summarized in Tables 4 and 5. The diffused glucose concentration is given in Table 4 and % of relative movement is given in Table 5. The movement of glucose from inside of membrane to external solution monitored and compared in Figures 4  and 5. Among the six samples only the 3e retains the glucose and shows minimum % of relative movement 41.06% over 180 minutes. All other samples show the highest % of relative movement in glucose diffusion from 30 to 180 minutes.

-Amylase Inhibitory Assay.
The results are given in Table 6. All samples show gradual increase in inhibition, where the sample concentration increased from 100 to 400 g/mL. Sample 3e shows maximum inhibition of 57% and sample 3d shows 23.55% of inhibition. All other samples show varying less significant inhibition.

-Glucosidase Inhibition Assay.
The results are shown in Table 7. Among the six samples, 3e shows maximum inhibitory activity of 60.25% at higher concentration   (400 g/mL). Sample 3d shows 23.01% of inhibition. All the other samples are less significant.

Molecular Docking Studies.
The synthesized molecules 3a-3f were constructed using VEGA ZZ molecular modeling package [35]. The obtained structures were then geometrically optimized using AM1 Hamiltonian in MOPAC software package [36]. The X-ray structure of pig -pancreatic -amylase (PDB 3L2M) and N-terminal human maltaseglucoamylase with Casuarina (PDB 3CTT) [37], obtained from Brookhaven Protein Data Bank, were used for docking calculations. Docking calculations were performed with AutoDock 4.0 [38]. The crystal structures were refined by removing water molecules and repeating coordinates. Hydrogen atoms were added and charges were assigned to the protein atoms using Kollman united atoms force field by using AutoDockTools-1.5.6. For docking calculations, Gasteiger partial atomic charges were added to the synthesized structures and all possible flexible torsion angles of the ligand were defined by using AUTOTORS. The structures were saved in a PDBQT format for AutoDock calculations. Values are mean ± SEM for groups of 3 observations. * < 0.01. * * < 0.05. AutoDock requires precalculated grid maps, one for each atom type present in the structure being docked. The auxiliary program AutoGrid generated the grid maps. Lennard-Jones parameters 12-10 and 12-6, implemented with the program, were used for modeling H-bonds and van der Waals interactions, respectively. The Lamarckian genetic algorithm method was applied for docking calculations using default parameters. AutoDock uses a semiempirical free energy force field to evaluate conformations during docking simulations. The optimized orientations represent possible binding modes of the ligand within the site: The first three terms are van der Waals, hydrogen bonding, and electrostatics, respectively. The term Δ tor is for rotation and translation and Δ desolv is for desolvation upon binding and the hydrophobic effect.
After docking, the 50 solutions were clustered into groups with RMS deviations lower than 1.0Å. The clusters were ranked by the lowest energy representative of each cluster.

Molecular Docking Study on -Amylase.
The binding site of the structures was not identified because of the absence of the crystal structure of the ligand, and a blind docking was performed for all the structures 3a-3f with the protein structure 3L2 M [39]. Two interacting binding sites were identified [37], one near the entrance of the central beta-barrel of the enzyme and the other near the Nterminal of the protein [39]. The docked structures showed binding energy in the range of −4.2 to −4.8 Kcal/mol. It was observed that the structure 3e exhibits high binding energy of −4.8 Kcal/mol and the structure 3d exhibits binding energy of −4.5 Kcal/mol, respectively. The binding energies (Δ BE ) and intermolecular energies (Δ intermol ) of the structures 3d and 3e obtained for -amylase are given in Table 8. When comparing 3d and 3e structures, 3e exhibits high binding energy. The structure 3e shows interaction with the residues Trp58, Trp59, Tyr62, Asp197, and Asp300 present in the binding site of -amylase. The amino acids Trp58 and Trp59 were the key residues interacting with all the structures. The docking studies revealed that the van der Waals, electrostatic, and desolvation energies play a key role in binding. It was observed that the Pyrimido ring was fitted well with binding site, via hydrogen bonds through the hydrogen atom of the amino group of 3e with carboxylate group of Asp300. The hydrophobic interactions were formed by Trp58, Trp59, Gln63, Val163, Asp300, His305, and Asp356 in 3e.

Molecular Docking Study on -Glucosidase.
All the binding modes of the structures 3a-3f were explored using the docking calculation in AutoDock. The binding site of the structures was identified using the crystal structure of Casuarina. The docked structures showed binding energy in the range of −6.3 to −6.6 Kcal/mol. The binding energies (Δ BE ) and intermolecular energies (Δ intermol ) of the structures 3d and 3e obtained for -glucosidase are given in Table 8. The structure 3e showed high binding energy when compared to 3d. Based on docking calculations, the synthesized compounds 3d and 3e show better inhibition of -glucosidase compared to -amylase which is in a good agreement with in vitro studies. The following key residues Asp203, Tyr299, Trp406, Asp443, Met444, Phe450, Lys480, Asp542, Phe575, and Gln603 present in the binding site of -glucosidase show noncovalent interactions with 3e. It was observed that the Pyrimido ring of 3e formed hydrogen bonds through the hydrogen atom of the amino group with carboxylate group of Asp203 at a distance of 1.9Å. The residues Tyr299, Trp406, Asp443, Met444, Phe450, Lys480, Asp542, Phe575, and Gln603 formed hydrophobic interactions with 3e.
In diabetes mellitus, control of postprandial plasma glucose level is critical in the early treatment [34]. Inhibition of enzymes involved in the metabolism of carbohydrates are one of the therapeutic approaches for reducing postprandial hyperglycemia. The inhibition by natural products is more safe than synthetic drugs.
-Amylase and -glucosidase are significant enzymes which cleave carbohydrates responsible for absorption of glucose in the blood stream. For this reason, synthetic drug such as acarbose, miglitol, and voglibose are some inhibitors which inhibit -amylase and -glucosidase [40]. However, these agents have their limitations, because they are nonspecific and produce serious side effects such as GI tract inflammation and to elevate diabetic complications [41].