5-Oxo-dihydropyranopyran derivatives as anti-proliferative agents; synthesis, biological evaluation, molecular docking, MD simulation, DFT, and in-silico pharmacokinetic studies

A series of ethyl 2-amino-7-methyl-5-oxo-4-phenyl-4,5-dihydropyrano[4,3-b]pyran-3-carboxylate derivatives (4a-j) bearing different substitutions on the C4-phenyl ring was synthesized. The anti-proliferative activity of all the synthesized compounds was assessed against two human cancer-cell lines, including SW-480 and MCF-7, by using MTT method. Derivatives 4g, 4i, and 4j, possessing 4-NO2, 4-Cl, and 3,4,5-(OCH3)3 substitutions, were found to be the most potent compounds against both cell lines. The obtained IC50 values for 4g, 4i, and 4j were 34.6, 35.9, and 38.6 μM against SW-480 cells and 42.6, 34.2, and 26.6 μM against MCF-7 cells, respectively. Evaluation of the free radical scavenging potential of the compounds against DPPH radicals showed the highest result for compound 4j with an EC50 value of 580 μM. Molecular docking studies revealed the compounds were well accommodated within the binding site of cyclin-dependent kinase-2 (CDK2) with binding energies comparable to those of DTQ (the co-crystallized inhibitor) and BMS-265246 (a well-known CDK2 inhibitor). Molecular dynamics simulation studies confirmed the interactions and stability of the 4g-CDK2 complex. All derivatives, except 4g, were predicted to comply with the drug-likeness rules. Compound 4j may be proposed as an anti-cancer lead candidate for further studies due to the promising findings from in-silico pharmacokinetic studies, such as high GI absorption, not being a P-gp substrate, and being a P-gp inhibitor. Density functional theory (DFT) analysis was performed at the B3LYP/6–311++G (d,p) level of theory to examine the reactivity or stability descriptors of 4d, 4g, 4i, and 4j derivatives. The highest value of energy gap between HOMO and LUMO and thermochemical parameters were obtained for 4i and 4j.

μM against SW-480 cells and 42.6, 34.2, and 26.6 μM against MCF-7 cells, respectively.Evaluation of the free radical scavenging potential of the compounds against DPPH radicals showed the highest result for compound 4j with an EC 50 value of 580 μM.Molecular docking studies revealed the compounds were well accommodated within the binding site of cyclindependent kinase-2 (CDK2) with binding energies comparable to those of DTQ (the cocrystallized inhibitor) and BMS-265246 (a well-known CDK2 inhibitor).Molecular dynamics simulation studies confirmed the interactions and stability of the 4g-CDK2 complex.All derivatives, except 4g, were predicted to comply with the drug-likeness rules.Compound 4j may be proposed as an anti-cancer lead candidate for further studies due to the promising findings from in-silico pharmacokinetic studies, such as high GI absorption, not being a P-gp substrate, and being a P-gp inhibitor.Density functional theory (DFT) analysis was performed at the B3LYP/ 6-311++G (d,p) level of theory to examine the reactivity or stability descriptors of 4d, 4g, 4i, and 4j derivatives.The highest value of energy gap between HOMO and LUMO and thermochemical parameters were obtained for 4i and 4j.

Introduction
Since the start of the twenty-first century, cancer has become a major leading cause of death globally [1].The incidence and mortality of cancer increases every year.The global cancer statistics indicate 19.3 million newly diagnosed cases and 10 million cancer-caused deaths worldwide in 2020 [2].Cancer prevalence is predicted to increase rapidly reaching 13.1 million cases in 2030 [3].The most significant attributes of cancer are uncontrolled proliferation, local invasion, and distant metastasis.Most cancer-related deaths are caused by tumor recurrence or distant metastasis after systemic antitumor therapy [4].Traditional treatments such as surgery, chemotherapy, radiotherapy, and immunotherapy are associated with several side effects that can lead to systemic adverse effects [5].The main disadvantages of systemic chemotherapy include low drug concentration in the tumor, rapid clearance from the circulation, and severe toxicity outside the tumor [6].Furthermore, some anti-cancer drugs are prone to resistance or have a short half-life due to rapid degradation by enzymes [6,7].Considering these limitations, scientists still endeavor to develop more effective small chemotherapeutic molecules with fewer side effects.
This study aimed to synthesize a series of substituted dihydropyranopyran derivatives (Fig. 1, 4a-j) as anti-cancer agents and investigate their cytotoxicity against two cancer cell lines.Molecular docking analysis is performed to evaluate the binding energies, modes, and compounds interactions within the active site of CDK2.Besides, the drug-likeness and pharmacokinetic properties of the derivatives are predicted in-silico.Finally, DFT analysis is carried out to determine the reactivity or stability as well as the nucleophilic and electrophilic sites of the studied compounds [25], using the B3LYP method with the 6-311++G (d,p) basis set.

Chemicals and apparatuses
The chemicals were purchased from Sigma-Aldrich (St. Louis, MO) and used without further purification.Reactions were monitored by thin layer chromatography (TLC) on MERCK precoated silica gel 60-F254 (0.5 mm) aluminum plates.Melting points were measured on a Kofler hot stage apparatus and were uncorrected.The Nuclear magnetic resonance (NMR) spectra were recorded on a Bruker 300 MHz NMR spectrometer with tetramethylsilane (TMS) as an internal standard. 1H NMR (300 MHz) and 13 C NMR (75 MHz) spectra were recorded in DMSO-d 6 .Chemical shifts (δ) were expressed in parts per million (ppm) unit, and J-coupling values were reported in Hertz (Hz).Mass spectra were determined with an Agilent spectrometer (Agilent Technologies 9575c inert MSD, USA).Infrared spectra (IR) were recorded on an FT-IR PerkinElmer Precisely system spectrophotometer (PerkinElmer, Waltham, MA) using the KBr disc technique.

In vitro anti-proliferative activity evaluation
Two cancer cell lines, including the human colorectal (SW-480) and the human breast (MCF-7) cells, were taken from the National Cell Bank of Iran (NCBI, Pasteur Institute, Tehran, Iran).All cells were cultured in RPMI-1640 medium containing 10 % fetal bovine serum (FBS) (Gibco Invitrogen Co., Scotland, UK), antibiotics (penicillin and streptomycin, 10 % v/v), at 37 • C in a humid atmosphere containing 5 % CO 2 .All designed compounds (4a-j) were evaluated by the standard 3-(4,5-dimethylthiazol-yl)-2,5-diphenyl-tetrazolium bromide (MTT) assay to check their anti-cancer activity.By using trypsin/EDTA 0.5 % solution, the cells were harvested and then seeded in 96-well microplates at a density of 1 × 10 4 and 8 × 10 3 cells per well, respectively, for MCF-7 and SW480 cell lines in S. Ranjbar et al. 100 μL of complete culture medium.After 24 h of incubation, the cells were treated with five different concentrations (1-500 μM) of cisplatin (as the positive control) and the synthesized derivatives in a triplicate manner.After 72 h of incubation, formazan crystals were obtained by replacing the media with 100 μL fresh MTT solution.The plates were incubated for 4 h at 37 • C. To dissolve the formazan crystals, the media was removed, 150 μL of DMSO was added, and the cells were incubated at 37 • C in the dark for 10 min.
Finally, a microplate ELISA reader was applied to record the absorbance of each well at 490 nm.To analyze the data, Excel 2016 and Curve Expert 1.4 [26] were used.The data were provided in terms of mean IC 50 value ± SD [27].

Free radical-scavenging activity evaluation
Radical-scavenging activity was determined by the 2,2-Diphenyl-1-picrylhydrazyl (DPPH) assay as described previously [12,28,29].A mixture of different concentrations of the compounds and DPPH methanolic solution (110 μM) was shaken in the dark at room temperature for 30 min.The mixture absorbance was measured at 517 nm.Quercetin was used as the positive control.After calculating the percentage of scavenging activities by the following equation, the EC 50 values were obtained from linear regression plots between the concentrations of the tested compounds and the percentage of the radical-scavenging activities.Each concentration was analyzed in three independent experiments conducted in triplicates.

Molecular docking study
Docking was performed by AutoDock 4.2 and AutoDock Tools 1.5.4 (ADT) [30].The X-ray crystal structure of CDK2 with 4-(3-hydroxyanilino)-6,7-dimethoxyquinazoline as the co-crystallized ligand (PDB ID: 1DI8) was taken from the Protein Data Bank (http://www.rcsb.org).Before docking, the innate ligand and water molecules were removed from 1DI8, then hydrogens were added, and nonpolar hydrogens were merged.Finally, Gasteiger charges were calculated for protein 1DI8.3D structures of the ligands were sketched and minimized using Avogadro software [31].PDBQT formats of the ligands were prepared by adding Gasteiger charges and setting the degree of torsions.The grid maps were constructed by considering a grid box of 50 × 50 × 50 dimensions with 0.375 Å spacing centered at x = − 9.052, y = 48.898,and z = 11.469Å.For the docking process, the macromolecule was considered to be rigid.Lamarckian genetic search algorithm was chosen, and the number of runs was set at 70.The validity of the docking procedure was confirmed using a co-crystallized inhibitor as the ligand and the above-mentioned protocol.

Molecular dynamics simulation
Molecular dynamics (MD) simulation analysis was performed to explore the dynamic behavior of the complexes.GROMACS 2016.3 software was utilized for simulations, and the CHARMM27 force field was employed to represent molecular interactions accurately [32].Ligand parameters were generated using the SwissParam web server.The system was solvated in a cubic periodic box, and counter ions were added for system neutralization.Following energy minimization, the system underwent equilibration in two successive phases: NVT (constant number of particles, volume, and temperature) and NPT (constant number of particles, pressure, and temperature).Subsequently, a production MD run was conducted with analyses based on a total of 100 ns MD trajectories.Root-mean-square deviation (RMSD), root-mean-square fluctuation (RMSF), and hydrogen bond analyses were performed to assess ligand stability and flexibility within the binding pocket over the simulation period.The molecular mechanics Poisson Boltzmann surface area (MM-PBSA) method was utilized to estimate binding free energies.All graphical representations were generated using Excel software.

Prediction of drug-likeness and pharmacokinetic properties
Drug-likeness profiling, GI absorption, and P-gp substrate potency of the synthesized compounds were predicted by the SwissADME (http://www.swissadme.ch)free web tool.The in-silico prediction of P-gp inhibitory activity, Blood-brain barrier (BBB) permeation, Caco 2 cell permeability, and MDCK cell permeability were carried out using preADMET (http://preadmet.qsarhub.com).

DFT analysis
The density functional theory (DFT) was carried out on Gaussian 09 at the B3LYP function coupled with the 6-311++G (d,p) basis set.The structural geometries were optimized at their ground state energies.Subsequently, energies of the highest occupied molecular orbitals (HOMO), lowest unoccupied molecular orbitals (LUMO), and energy gap between HOMO and LUMO were obtained for 4g, 4i, 4j, and 4d.Moreover, thermochemical parameters were determined by using HOMO and LUMO energies.The Theoretical FT-IR spectra were also calculated at their optimized geometries through DFT/6-311++G (d,p) functions.

Synthesis
The general procedure for the synthesis of the target dihydropyranopyran derivatives is shown in Scheme 1.A mixture of 4-S.Ranjbar et al. hydroxy-6-methyl-2-pyrone (1), corresponding benzaldehyde (2a-j), and ethyl cyanoacetate (3) in the presence of ammonium acetate was refluxed in water and ethanol for 24 h.The final pure products (4a-j) were obtained after washing with a suitable solvent or, in some cases, after applying TLC.The structures of the synthesized compounds were confirmed by 1 H NMR, 13 C NMR, MS, and FT-IR analyses.

Anti-proliferative activity of the synthesized compounds
The anti-proliferative activity of all the compounds was investigated against SW-480 and MCF-7 cancer cell lines by using MTT assay.Cisplatin was used as the positive control.The concentration required to inhibit the cancer cells' growth by 50 % (IC 50 ) was estimated for each derivative (Table 1).Compounds 4g, 4i, and 4j (bearing 4-NO 2 , 4-Cl, and 3,4,5-(OCH 3 ) 3 , respectively) with IC 50 values of 34.6, 35.9, and 38.6 μM, respectively, against SW-480 cell line, and 42.6, 34.2, and 26.6 μM, against MCF-7 cells showed the highest anti-proliferative effects.Compounds 4a, 4d, and 4h had the lowest anti-proliferative activities against SW-480 with IC 50 values of 90.5, 87.2, and 67.2 μM, respectively.Derivatives 4d and 4e were the least active compounds against MCF-7 cells with IC 50 values of 104.9 and 71.0 μM.The rest of the compounds exhibited moderate anti-proliferative activities with an IC 50 range of 41.2-54.3μM.
According to the results, a brief structure-activity relationship (SAR) can be proposed.It seems that substitution on the C 4 -phenyl ring improved the anti-proliferative activity of the ethyl 2-amino-7-methyl-5-oxo-4,5-dihydropyrano[4,3-b]pyran-3-carboxylate derivatives; as most of the compounds had superior activities to derivative 4a.Exceptions were 4d, 4e, and 4h (bearing 3-Cl, 4-CN, and 4-Br moieties on the phenyl ring, respectively), which showed IC 50 values higher than 4a.Placing an electron-withdrawing chlorine atom at the meta position of the phenyl ring (as in 4d) diminished the activity; however, substituting electron-donating groups at this position (as in 4b bearing 3-OH substitution) enhanced the activity.Accordingly, introducing three electron-donating methoxy groups at the 3, 4, and 5 positions of the phenyl ring caused compound 4j to be one of the most potent derivatives.The nature of the substituents introduced to the para position of the C 4 -phenyl had a considerable effect on the anti-proliferative potency of the dihydropyranopyran derivatives.It seems that the insertion of electron-withdrawing moieties at this position would reduce activity.Compound 4i showed better activity than 4h and 4e (bearing strong electron-withdrawing 4-CN and 4-CF 3 groups) due to the reduced electron-withdrawing capacity of the chlorine atom at the para position.In compound 4g, the insertion of a nitro group at the para position as a hydrogen bond acceptor moiety, led to a noticeable increase in activity compared with 4h and 4e.

Molecular docking analysis
Molecular docking is a widely used technique to assess in-silico the affinity and inhibitory potential of compounds showing anticancer activity against a specific target [27,33,34].Cyclin-dependent kinases (CDKs) are a family of proteins that contribute to cell proliferation by regulating the progression of cell cycle and transcription [35,36].A literature review revealed that CKD2 can be proposed as a valid target for the anti-cancer pyran derivatives [21][22][23][24].CKD inhibitors compete with ATP to bind at the kinase site, causing suppression of CDK hyperactivation through inhibition of kinase phosphorylation and, consequently, preventing extreme cell proliferation [37].Therefore, molecular docking analysis was carried out to predict the binding affinity and interactions of the dihydropyranopyran derivatives in this study (4a-j) into the ATP binding site of CDK2.Scheme 1. Synthetic route for the preparation of 5-oxo-4,5-dihydropyrano[4,3-b]pyran derivatives.

S. Ranjbar et al.
The docking procedure was initially validated using a self-docking approach.The co-crystallized ligand (DTQ) was re-docked into the binding site of CDK2 (PDB code: 1DI8).The RMSD between the best pose of the co-crystallized ligand docked into the binding site of CDK2 and the one in the crystal structure was found 1.46 Å.This confirmed the validity of the docking procedure.The calculated binding free energies (ΔG) and interaction details for the target compounds (4a-j), DTQ and BMS-265246 (a well-known CDK2 inhibitor), are presented in Table 2.All derivatives were well accommodated within the active site of CDK2 with binding energy values ranging from − 7.52 to − 8.72 kcal/mol, which was comparable to the binding energies of DTQ (− 8.00 kcal/mol) and BMS-265246 (− 7.76 kcal/mol).The 3D binding orientations of the derivatives are illustrated in Fig. 2. The compounds had different binding modes.It can be concluded that the nature and position of the substitutions on the C 4 -phenyl ring determined the orientations of the derivatives in the binding site of CDK2.The ligand-protein interactions are depicted in Fig. 3.As shown in Fig. 2a, compounds 4a and 4b had the same binding orientations.The NH 2 moieties established two hydrogen bonds with GLU81 residue, and the oxygen atom in the dihydropyran part of the compounds made an additional hydrogen bonding interaction with LEU83.The pyranon nucleus formed hydrophobic interactions with LEU134, ILE10, GLN85, and ASP86 amino acids, while the C 4 -phenyl ring contributed to establishing hydrophobic interactions with ALA144, ASN132, GLN131, LEU133, and VAL18 residues.Moreover, the ethyl of carboxylate groups occupied the hydrophobic pocket surrounded by PHE80, ALA144, and VAL64 residues.In the case of compound 4b, the 3-OH group provided two additional hydrogen bonding interactions with LYS33 and ASP145.According to the docking results, 4i, 4g, 4c, and 4e were oriented similarly in the ATP binding site of CDK2 (Fig. 2b).The pyranon ring contributed to the formation of hydrophobic interactions with ILE10, GLY11, GLN131, ASN132, and LEU134 residues.The phenyl ring was surrounded by lipophilic residues, including ALA144, LEU134, ALA31, and VAL18.A key hydrogen bonding interaction between NH 2 and LEU83 was observed in the mentioned compounds.The 4e and 4g derivatives made another hydrogen bond with LEU83 through the carboxylate substitution.Moreover, 4e and 4g formed a hydrogen bond with amino acid residue ASP145 through the phenyl ring substitution (4-CN and 4-NO 2 ).Compound 4g formed a more stable complex via establishing an additional hydrogen bond with LYS33 through the 4-NO 2 substitution.In compounds 4f and 4h (Fig. 2c), carboxylate and amino substitutions were involved in hydrogen bonding interactions with LEU83 and GLU81, respectively.The pyranon and its methyl substitution occupied the hydrophobic pocket comprising PHE80, ALA144, ALA31, VAL64, LEU134, LYS33, and PHE80.The phenyl ring contributed to hydrophobic interactions with the nearby residues, including ILE10, GLY11, and VAL18.Furthermore, the 4-CF 3 substitution in compound 4h was involved in halogen bonding interaction with ILE10, GLY11, and GLU12 residues.Compounds 4d and 4j were oriented differently in the CDK2 active site (Fig. 2d).In the case of compound 4d, two hydrogen bonds were observed between the amino group and the two residues, ASN132 and GLN131.The phenyl ring showed lipophilic interaction with VAL18, ALA31, VAL64, GLU81, and PHE82, while the pyranon moiety was accommodated in the pocket comprising ALA144, LYS33, VAL18, PHE80, LEU148, and ASP145 amino acids through hydrophobic interactions.3-Cl and ethyl carboxylate groups established hydrophobic interaction with ILE10.Finally, compound 4j made a stable ligand-CDK2 complex by establishing hydrogen bonding, electrostatic, and hydrophobic interactions.This derivative formed three hydrogen bonds; with amino acid residue ASP145 through its amino and carboxylate groups and amino acid LYS33 through the oxygen atom of the dihydropyran ring.The pyranon ring contributed to an electrostatic interaction with ASP145 as well as some hydrophobic interactions with ALA144, GLN131, and ASN132.The trimethoxyphenyl ring was surrounded by hydrophobic residues including VAL64, LEU83, LEU134, ALA31, ALA144, and ILE10.
The docking results exhibited that stabilization of DTQ in the active site of CDK2 resulted from key hydrogen bonds with residues LEU83 and ASP145 at the distances of 2.08 and 2.28 Å, respectively, as well as an electrostatic interaction with ASP145 at a distance of     4.56 Å, and some hydrophobic interactions with ILE10, VAL18, ALA31, LYS33, AVAL64, GLU81, PHE82, LEU83, HIS84, LEU134, and ALA144.All the active derivatives established hydrogen bonds with one or both LEU83 and ASP145 residues via the amine group, the carboxylate moiety, the oxygen atom of the dihydropyran ring, or the substitutions on the C 4 -phenyl ring.The least potent compound 4d, did not form any hydrogen bonding with the mentioned amino acids.Moreover, in the target compounds, ethyl of the carboxylate group, pyranon ring, and the C 4 -phenyl participated in establishing hydrophobic interactions with the same residues as DTQ did.The substitutions on the C 4 -phenyl ring had a key role in the orientations and interactions of the derivatives.For example, in the case of 4j, the three methoxy substitutions on the C 4 -phenyl ring provided some strong hydrophobic interactions with VAL64, LEU83, LEU134, ALA31, and ILE10; while, in the case of compound 4g, the nitro substitution formed tough hydrogen bonding interactions with LYS33 and ASP145.

MD simulations
To evaluate the stability of compound 4j and DTQ within the CDK2 active site, MD simulations were conducted.The RMSD analysis helped to study the degree of deviation from the initial structure during the simulation time.Fig. 4a presents the average backbone RMSD values for compound 4j and DTQ complex structures over the 100 ns simulation period.Compound 4j exhibited an average RMSD value of 2.5 Å, with fluctuations ranging from 1.0 to 3.7 Å. DTQ complex displayed a lower average RMSD value of 1.9 Å. Notably, fluctuations in the RMSD value for the DTQ complex remained within the range of 1.0-2.5 Å.The RMSF values explained the structural integrity of CDK2 and the flexibility of amino acid residues bound to compound 4j and DTQ during the 100 ns MD simulation period (Fig. 4b).Significant fluctuations were observed in the amino acid residues located at the terminal end sections.Moreover, 4j and DTQ complexes displayed a similar RMSF pattern, ranging from 0.5 to 6 Å, with average values of 1.1 and 0.9 Å, respectively.
To further evaluate the stability of the protein-ligand complex, hydrogen bonding interactions were analyzed throughout the MD simulation.The calculation and analysis of hydrogen bonding interactions were carried out by VMD (Visual Molecular Dynamics), version 1.9.3 [38], and the H-bond module in GROMACS.The default hydrogen bonding criteria were set as d ≤ 3.5 Å and θ (X-H-A)≥ 20 • for the calculations.The number of hydrogen bonds formed by 4j and DTQ within the active site of CDK2 during the simulation is depicted in Fig. 5.The findings indicated that he DTQ exhibited three hydrogen bonds throughout the simulation (Fig. 5a), whereas the 4j formed four hydrogen bonds (Fig. 5b).Additionally, the occupancy percentages, which indicate the stability and the lifetime of each hydrogen bond are presented in Table 3.According to these results, 4j established the most stable hydrogen bonds with ASP145 and LYS33, and DTQ formed hydrogen bonds with LEU83 and ASP145.These are consistent with the docking result, suggesting that compound 4j maintains the same binding mode with the CDK2 active site during simulation.
The molecular mechanics Poisson Boltzmann surface area (MMPBSA) method was used to assess the binding free energy of compound 4j and the native ligand DTQ.Decomposition energy analysis per residue was calculated to determine the contribution of each residue to the binding energy of the ligand in the CDK2 complex.The binding energy terms and their contribution to the total binding energy of 4j and DTQ at the active site of the CDK2 are listed in Table 4. Compound 4j exhibited a binding energy (− 87.14 ± 3.01 kJ/mol) comparable to DTQ (− 81.46 ± 2.72 kJ/mol).The favorable energy term for 4j binding energy was to be driven by the van der Waals (VDW) energy rather than the electrostatic energy.This justifies the low number of hydrogen bonds and their low occupancies obtained in the hydrogen bonding analysis.The hydrogen bonding represented in the electrostatic energy term was more prominent for compound 4j (− 39.255 ± 2.317 kJ/mol) than that observed for DTQ (− 23.630 ± 2.135 kJ/mol).This finding corroborates our earlier observations regarding the higher number of hydrogen bonds formed by compound 4j than DTQ.Moreover, 4j showed the VDW energy value of − 169.51 ± 1.93 kJ/mol, which was more negative than DTQ (− 138.15 ± 1.88).This is in line with our finding in docking studies in which 4j mainly established hydrophobic interactions with the hydrophobic residues of the active site through trimethoxyphenyl, pyranon, and ethyl moieties.
The per-residue decomposition energy analysis graph for the DTQ-CDK2 and 4j-CDK2 complexes is depicted in Fig. 6(a and b).Accordingly, ILE10, VAL17, VAL18, ALA31, LYS33, PHE80, PHE82, GLN131, ASN132, LUE134, ALA144, and ASP145 of CDK2 played pivotal roles in the stabilization of 4j bound to the receptor.This is consistent with the results of molecular docking and hydrogen bond analysis.
The estimated pharmacokinetic profiles of 4a-j are listed in Table 5.Moreover, all compounds were predicted to have high GI absorption, except for 4g, which showed low GI absorption, indicating that other compounds are well orally absorbed.All the derivatives presented low to moderate blood-brain barrier (BBB) permeation; consequently, they are less likely to cause neurotoxicity (4g, 4h, and 4e have the lowest BBB permeation).All the derivatives showed medium permeability in human colorectal carcinoma cells (Caco-2) and low permeability in Madin-Darby Canine Kidney cells (MDCK).All derivatives, except 4g, are not P-glycoprotein (Pgp) substrates [45].P-gp has been identified as the most important efflux transporter responsible for multidrug resistance (MDR) of cancer cells to chemotherapeutic agents pumping anti-cancer drugs outside the cell.Therefore, the anti-cancer potency of the  compounds might not be affected by P-gp.Compounds 4g, 4h, and 4j were predicted to have an inhibitory effect on the P-gp pump.Hence, co-administration of these cytotoxic agents with common anti-cancer drugs might reverse MDR.

DFT approach
Molecular orbitals such as HOMO and LUMO, as well as the energy gap between them (ΔEgap), determine the chemical reaction site and molecular kinetic stability.HOMO energy indicates the capacity of the molecule to donate electrons, and LUMO energy indicates the capacity of the molecule to accept electrons.HOMO and LUMO energies using B3LYP/6-311++G (d,p) were measured and are presented in Fig. 7.The energy gaps between HOMO and LUMO were obtained at 4.293, 4.293, 3.834, and 3.744 eV for 4i, 4j, 4g, and 4d, respectively.A high energy gap indicates that the ligand is more stable, while a low energy gap indicates that the ligand is more reactive.In many cases, the energy gap confirms the biological behavior of a molecule.The energy gaps for 4i, 4j, and 4g derivatives were lower than that of derivative 4d, which indicated more stability of these derivatives.As can be seen, the LUMO of 4i derivative is concentrated on the whole molecule, and the HOMO of this compound is distributed in most parts of the molecule except the pyranon ring (Fig. 7a).For compound 4j, LUMO is located on the entire molecule except the ethyl carboxylate group and HOMO is on the dihydropyranopyran core, NH 2 , and carboxylate moiety (Fig. 7b).For compound 4g, LUMO is placed on the entire compound except the 4-nitrophenyl ring, while HOMO is placed only on this ring (Fig. 7c).In the case of 4d, HOMO and LUMO are distributed throughout the molecule (Fig. 7d).
Electrostatic surface potential (ESP) describes the electron density around a molecule and is a useful indicator for determining nucleophilic and electrophilic sites.The diagrams for the compounds 4i, 4j, 4g, and 4d are plotted in Fig. 8(a-d).The negative regions represent the electrophilic sites (red color), while the positive regions indicate the nucleophilic centers (blue color).
Theoretical chemistry can help to estimate the reactivity or stability of chemical species.Quantum chemical descriptors predict the chemical reactivity of molecules, the analysis of reactions, and the location of reactions in molecules.Thermochemical parameters such as total energy (E tot ), enthalpy energy (H), Gibbs energy (G), and entropy energy (S) were measured based on the B3LYP/ 6-311++G (d,p) level of theory.Moreover, hardness (η), softness (σ), and electron affinity (A) were determined using HOMO and LUMO energies.The values of chemical reactivity indices for derivatives 4i, 4g, 4j, and 4d are presented in Table 6.According to the results, the energy values of 4i, 4j, and 4g derivatives are more than 4d, so these compounds are more stable.Additionally, the energy value of the 4i derivative shows a higher stability of this molecule than other compounds.It is noticed that molecules with a larger   energy gap are harder and have higher kinetic stability because of resistance to electron cloud deformation.Considering that 4i, 4g, and 4j derivatives have a larger energy gap, the hardness parameter of these compounds is expected to be higher than that of compound 4d.The value of electron affinity for compound 4d is calculated to be higher than other compounds, indicating the greater tendency of this compound to accept electrons, which is consistent with other obtained results.The IR peaks were theoretically calculated at the B3LYP/6-311++G (d,p) level of theory for all derivatives, as shown in Fig. 9.By using the IR spectrum, aromatic and aliphatic C-H, C --O, and NH stretching vibrations can be seen clearly, which help to identify compounds' structures.C --O stretching vibrations in the 1778.22, 1776.45, 1786.66, and 1779.31cm − 1 regions were observed for 4i, 4g, 4j, and 4d compounds, respectively.Furthermore, the C-H aromatic stretching vibrations were at ~3000 cm − 1 region, and aliphatic vibrations were visible around ~2900.NH stretching vibrations were observed in the range 3300-3500 cm − 1 .The results indicated that theoretical and experimental IR spectra were matched.

Table 5
The drug-likeness and pharmacokinetic properties predicted results for 4a-j.

Fig. 3 .
Fig. 3. Interactions of the target compounds (4a-j) in the binding site of CDK2.

S
.Ranjbar et al.

Fig. 5 .
Fig. 5.The number of hydrogen bonding interactions of (a) DTQ and (b) 4j within the active site of CDK2 during 100 ns simulation.

S
. Ranjbar et al. cell lines SW-480 and MCF-7.Compound 4j, bearing a 3,4,5-trimethoxy phenyl at the C 4 position, was among the most potent derivatives with IC 50 values of 38.6 and 26.6 μM against SW-480 and MCF-7 cells, respectively.Moreover, 4j proved to be a DPPH radical scavenger with an EC 50 value of 580 μM.Molecular docking studies revealed that all the compounds could bind to the CCDK2 active site with good affinity.Excellent drug-likeness profiling and promising pharmacokinetic properties were predicted for almost all the dihydropyranopyran derivatives by in-silico studies.The findings imply that 4j might be considered a potential lead molecule for additional research in the field of anti-cancer drug development.

Table 1
Chemical structures, anti-proliferative, and radical scavenging activities of the synthesized compounds.
a Values represent means ± SD of three independent experiments.S.Ranjbar et al.

Table 2
Docking results of the target derivatives (4a-j) within the binding site of CDK2.

Table 3
Hydrogen bond analysis of compound 4j and DTQ within the active site of CDK2.

Table 4
Binding affinities and individual energy components resulting from MM-PBSA analysis of 4j and DTQ complexes.
a S.Ranjbar et al.