Synthesis, anticancer activity, molecular docking and molecular dynamics studies of some pyrazole–chalcone hybrids

Abstract Four new hybrid compounds (H1–H4) bearing pyrazole (S1 and S2) and chalcone (P1 and P2) fragments were synthesized and characterized. Compounds were assayed for their ability to inhibit the proliferation of human lung (A549) and colon (Caco-2) cancer cell lines. Besides, toxicity against normal cells was determined using the human umbilical vein endothelial cells (HUVEC). In silico molecular docking, molecular dynamics (MD) simulation and absorption, distribution, metabolism, excretion, and toxicity (ADMET) studies were carried out to predict the binding modes, protein stability, drug-likeness and toxicity of the reported compounds. The in vitro anticancer activity of the tested compounds revealed dose-dependent cell-specific cytotoxicity. In silico studies revealed that the compounds have a good binding affinity, possess appropriate drug-likeness properties and have low toxicity profiles. Communicated by Ramaswamy H. Sarma


Introduction
Pyrazole-based compounds have been a topic of numerous research due to their multi-dimensional applications, especially in medicinal chemistry (Fustero et al., 2011;Khan et al., 2016).This high interest is essentially due to their ability to interact with the bio-receptors and mimic biomolecules (Bowers & Sugiyama, 1993).Researchers across the globe have continuously pursued new pyrazole derivatives as anticancer, antiviral, antimicrobial and others with great success (Li et al., 2022).This is further evident by the approval of various small pyrazole-based molecules by US-FDA in the last few years (Becerra et al., 2022).Besides, several groups attempted the synthesis of bioactive compounds by merging pyrazole with other bioactive fragments (Fustero et al., 2011).The technique, referred to as molecular hybridization, is a powerful way to merge features and properties of two promising scaffolds to obtain new drug candidates (Soltan et al., 2021).Especially chalcones-pyrazole hybrids have gained a particular interest from the medicinal chemistry point of view (Gao et al., 2020).Several examples of pyrazole-chalcone hybrids have been reported with broad-ranging bioactivity.For instance, Bandgar et al. (2009) noted that pyrazole-chalcone hybrids (i, Figure 1) exhibit anti-inflammatory, antioxidant and antimicrobial properties with low toxicity.It was noted that the electronic nature of the functional group has a remarkable effect on biological activity, especially antibacterial.When the C(3)-position of the N-phenyl pyrazole was functionalized with different heteroaromatics and C(4) with substituted chalcone, it yielded compounds of series ii (Figure 1) with cytotoxicity in micromolar range towards a number of cancerous cells (Hawash et al., 2017).Flow cytometry analysis revealed that the compounds were able to arrest G2/M phase of the cell cycle and induce apoptosis via inhibition of kinases.It is interesting to note that a minor change in the functional group led to remarkable change in activity of this class of molecules.For instance, among a number of compounds assessed in the series, compound iii and iv (Figure 1) was found to be highly active against cancer cell lines (Insuasty et al., 2010;Rai et al., 2015).However, their structural counterpart remains inactive.
Motivated by the findings, we report the synthesis, characterization, biological and in silico evolution of four new hybrid compounds (H1-H4) bearing pyrazole (S1 and S2) and chalcone (P1 and P2) fragments (v, Figure 1).The anticancer activity was assessed against lung (A549) and colon (Caco-2) cancer cell lines in vitro.Compounds were further evaluated for their biological safety profiles using the human umbilical vein endothelial cells (HUVEC).Drug-likeness properties and docking studies were carried out to underpin their suitability as a drug and mechanism of action.Furthermore, we analyzed the conformational stability of the docked complexes using molecular dynamics (MD) simulations with the help of various parameters such as root mean square deviation (RMSD), root mean square fluctuation (RMSF), the radius of gyration (Rg) and solvent accessible surface area (SASA).

Materials and methods
Unless stated otherwise, all chemicals were obtained from Sigma Aldrich and used as received.Column chromatography was performed on silica gel using dichloromethane/hexane as eluent.Precoated sheets (silica gel 60 F 254 , Merck Germany) were used for thin-layer chromatography (TLC), and the spots were visualized under UV light.Attenuatedtotal-reflectance IR spectra were recorded on pure samples Schimadzu IRSpirit-T spectrometer.Liquid chromatography/mass spectrometry (LC/MS) was performed on an Agilent LC/MS instrument (1260 Infinity II) equipped with a reversephase C 18 column (2.7 mm particle size, 3.0 � 100 mm), electrospray (ESI) mass spectrometry detector, and photodiode array detector reverse-phase, electrospray (ESI) mass spectrometry detector, and photodiode array detector.Nuclear magnetic resonance (NMR) was recorded on Bruker Spectrospin DPX spectrometer (125/500 MHz).Tetramethylsilane (TMS) was used as an internal standard.The following abbreviations were used in reporting spectra: s ¼ singlet, d ¼ doublet, dd ¼ double doublet, m ¼ multiplet.
The obtained solid was washed with cold ethanol and used without further purification.Pyrazole aldehydes (S1 or S2) were obtained via Vilsmeier Haack formylation reaction of P1 and P2 adopting the standard protocol (Hawash et al., 2017).

Molecular docking studies
AutoDock4 (version 4.2.6) was used for the molecular docking (Trott & Olson, 2010).The crystal structure of the receptor (2VTO) was downloaded from the Brookhaven Protein Data Bank.(http://www.rcsb.org).Input files of the ligands (compounds) and receptor (2VTO) were converted from .pdb to .pdbqtformat.Blind docking was performed using the grid size of 96 � 76 � 102 Å and space of 0.375 Å. Default parameters for Lamarckian genetic algorithm (LGA) were used to generate the best molecular conformation of the ligands.After completion of run, output files (.dlg format) were analyzed with PyMol.(DeLano, 2002).

Molecular dynamics (MD) simulation studies
Final compound (H2) was selected for the molecular dynamics (MD) simulation studies.MD simulations were performed using GROMACS Version 5.18.3.package (Abraham et al., 2015).The topology of cyclin-dependent kinase-2 (CDK2) was generated by using GROMOS9643a1 force field (Van Der Spoel et al., 2005) while the molecular topologies and coordinate files were generated using PRODRG server (Sch€ uttelkopf & Van Aalten, 2004).Systems were solvated using a simple point charge model (SPC/E) in a cubic box.0.15 M counter ions (Na þ /Cl À ) were added to neutralize the system.The energy minimization of all the neutralized systems was performed using the steepest descent and conjugate gradients (50,000 steps for each).The regulation of volume (NVT) and pressure (NPT) was run for system equilibration.The NVT ensemble was employed at constant temperature of 300K and constant pressure of 1 bar.The SHAKE algorithm was used to confine the H atoms at their equilibrium distances, and periodic boundary conditions.Moreover, the Particle Mesh Ewald (PME) method used to define longrange electrostatic forces (Abraham et al., 2015).The cut-offs for Van der Waals and columbic interactions were set at 1.0 nm.LINC algorithm was used to constrain the bonds and angles.Using the NPT ensemble, production runs were performed for the period of 100 ns, with time integration.The energy, velocity, and trajectory were updated at the time interval of 10 ps.The Ca-atom deviations of the protein were calculated using root mean square deviations (RMSD).The relative fluctuations of each amino acid were defined with root mean square fluctuations (RMSF).To measure the compactness of a given molecule radius of gyrations (Rg) was determined, and the solvent accessible surface area (SASA) was employed to underpin the electrostatic contributions of molecular solvation (Ahamad et al., 2018(Ahamad et al., , 2019)).

Cell lines and culture condition
Human lung adenocarcinoma (A549), colorectal adenocarcinoma (Caco-2), and human umbilical vein endothelial (HUVEC) cell lines were grown as a monolayer culture in DMEM media supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin solution at 37 � C with 5% CO 2 in a humidified chamber.

Cell viability assay
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction assay was carried out as described previously with slight modification (Kamran & Gude, 2012).5000 cells were seeded in 96 well plate and incubated at 37 � C overnight.Compounds were dissolved in dimethyl sulfoxide (DMSO) and then diluted with a 10% FBS-containing medium.Cells were treated with different concentrations (10, 25, 50 and 75 mg/mL) of each compound for 24 h.Cells treated with the DMSO (0.2% in media) alone served as the vehicle control.After 24 h, the medium was removed, and cells were washed with PBS.MTT solution (1 mg/mL) in the complete medium was added to each well.Cells were reincubated at 37 � C for four hours.Following the removal of the medium, 100 ml of DMSO was added to each well to dissolve the formazan crystals.The plate was maintained at 37 � C for 10 min.Absorbance was measured at 540 nm using a microplate reader (BIORAD-680).

Statistical analysis
One-way ANOVA was performed using Graph Pad Prism-5 for statistical analysis of the data.The results were provided as the mean standard deviation of at least three independent measurements (nonsignificant, ns; p 0.05, � p 0.01, � p 0.001 versus the untreated control).

Synthesis
Several researchers investigated the chemistry and structural features of pyrazole-chalcone hybrids recently (Fustero et al., 2011).The synthetic route adopted in this work is depicted in Scheme 1. Phenylhydrazones P1 and P2 were prepared by the condensation of 2-chlorophenylhydrazine and substituted acetophenones in absolute ethanol.The products were purified by column chromatography.Subsequently, S1 and S2 was obtained by POCl 3 /DMF mediated Vilsmeier-Haack reaction.Finally, hybrid compounds H1-H4 were obtained by base (40% NaOH) catalyzed pyrazolealdehyde-acetophenone condensation reaction.The structures of all the synthesized compounds were established by MS, 1 H-NMR, and 13 C-NMR spectroscopy (vide-infra).

Nuclear magnetic resonance (NMR) spectroscopy
The chemical structure of the compounds was further confirmed by using proton ( 1 H) and carbon ( 13 C) NMR spectroscopy (Figures S9-S15

Mass spectrometry (MS)
Mass spectra data of the compounds were collected, and the results are depicted in Figure S16-S20 (supporting information).Compound P2 showed a molecular ion peak at m/z 332.99 (calc.323.62) attributed to (M) þ peak.Similarly, compound S1 and S2, (M) þ peaks at m/z 287.7 (calc.282.73) and 361.97 (calc.361.6) confirms the formation of the products.In the MS spectra of final compounds, the presence of a molecular ion as well as base peak conforms the molecular structure of the final products.

Cytotoxicity assay
The antiproliferative potential of several natural and synthetic compounds have been tested on human umbilical vein endothelial cell (HUVEC) in the past (Medina-Leyte et al., 2020;Rhim et al., 1998;Singh et al., 1996).Since HUVEC serve as an excellent non-cancerous cell model, we determined the cytotoxicity of compounds against HUVEC by MTT assay and the result is given in Figure 2a, b.All compounds exhibited cytotoxicity in a concentration-dependent manner.For instance, compound S1 showed a viability of 62.6% at 10 lg/mL which reduced to 42.2, 21.9 and 16.7% at 25, 50 and 75 lg/mL, respectively.Similar toxicity profiles showed by the hybrids H1-H4 with viability ranging between 61.85 and 69.0% at 10 lg/mL, 53.1-64.1% at 25 lg/mL, 45.4-52.4% at 50 lg/mL, and 34.4-41.2%at 75 lg/mL.Among intermediate compounds, S1 has an IC 50 values of 19.1 lg/mL while it was insignificant for S2 (data not included).Compounds H1-H4 showed comparatively lower toxic effects in the noncancerogenic HUVECs and was in the order: H4 (40.9 lg/mL) < H2 (41.9 lg/mL) < H1 (47.0 lg/mL) < H3 (54.3 lg/mL).This result is in line with the earlier reports where it was demonstrated that the toxicity of the N-phenylpyrazole-chalcone derivatives depends on the functionalities of the substituents present at 3 and 4-position of the pyrazole cores (Hawash et al., 2017;Insuasty et al., 2010).
The anticancer activity of the intermediate, final pyrazole-chalcone derivatives and doxorubicin (as a positive control) was tested against human lung (A549) and colorectal (Caco-2) adenocarcinoma cell lines.The results of the assay are depicted in Figure 3.As observed against the non-cancerous HUVECs, the compounds also exhibited a dose-dependent anticancer activity against cancerous cells.
The compounds showed anti-tumour activity at lower concentrations (10 lg/mL).But at higher concentrations (from 25 to 75 lg/mL), intermediate compound S1 showed higher cytotoxic activity than other compounds against A549.Against the same cell line, the hybrid compounds' activity order was H1 > H4 > H2 > H3 at all concentrations.Contrarily, all the tested compounds exhibit almost similar activity on Caco-2 cell line, with S1 being the most active while H2 being the least at 25 lg/mL.Potent activity of pyrazole-chalcone derivatives against leukemia, renal and non-small lung cancer cell lines are already known (Hawash et al., 2017;Insuasty et al., 2010).Isloor and coworkers (Rai et al., 2015) noted that the anticancer activity of pyrazole-chalcone derivatives bearing chalcone with halogen atom is higher than the others on breast (MCF-7) and cervical (HeLa) cancer cell lines.Besides, the position of halogen substituent on the ortho, meta or para position (aromatic present at C-3 position) also play a crucial role in the activity (Bandgar et al., 2009).

Molecular docking studies
Cyclin-dependent kinases (CDKs) play a vital role in the cell cycle and are over-expressed in many types of cancer, thereby serving as potential targets for developing new anticancer agents (Fischer & Lane, 2000).Substituted pyrazolebased compounds are known to bind and inhibit CDKs (Huang et al., 2007;Reddy et al., 2016;Sun et al., 2013;Wyatt et al., 2008), including CDK2 in cancer cells (Wang et al., 2021).Considering this, we carried out a molecular docking study using the CDK receptor (PDB: 2VTO) and the intermediate (S1 & S2) as well as the final compounds (H1-H4).Results of the docking study are depicted in Table 1 and Figure 4.As reported, a CDK2 inhibitor may interact with the receptor via various vital residues, including Ile10, Phe82, Asp86 and Leu134 (hydrophobic pocket), Phe80 (gatekeeper residue), Asp145 (DFG motif), Lys89, Asp86 (solvent accessible region), Glu81 and Leu83 (hinge region; Wyatt et al., 2008).As can be seen, among intermediate compounds, S1 (binding affinity ¼ À 8.9 kcal/mol) showed slightly higher affinity than S2 (binding affinity ¼ À 8.7 kcal/mol) to the receptor in the hinge region (Leu83).In the case of hybrid compounds, H1 showed the greatest binding affinity (À 10.02 kcal/mol) followed by H3 (À 9.80 kcal/mol), H4 (À 9.41 kcal/mol) and H2 (À 9.09 kcal/mol).This interaction pattern shows that the hybrids exhibit higher affinity towards the receptor than the intermediate compounds, which can be ascribed to the presence of additional functionalities in the final compounds.

Molecular dynamics (MD) simulation studies
To further shed light on the stability and ligand-protein interactions, molecular dynamics (MD) simulation study was performed.One of the final complexes with compound (H2) was selected for 500 ns simulations.The outputs (RMSD, RMSF, Rg and SASA) were evaluated as the function of time (Table 2).

RMSD and RMSF.
The docked complexes were subjected to RMSD analysis to assess the residual flexibility of CDK2.It was noted that the native protein CDK2 exhibits higher average RMSD value (�0.43 nm).Besides, native protein equilibrates (� 0.4 nm) at 50 ns followed by higher fluctuation (0.32-5.2 nm) between 100 and 300 ns.On the other hand, CDK2 and H2 complex reached equilibrium within 0.3 ns with an average RMSD value of 0.32 nm.The RMS deviation of Ca-atoms increased (0.35 nm) between 75 and 190 ns and thereafter remain stable (Figure 5a).Overall, the results revealed higher RMSD values with native and complex with H2, revealing a stable binding under the given  Table 3. Representative drug properties data with simple color and shape examples (Ritchie et al., 2011).
Small horizontal bars and light color of the properties indicate more drug-likeness.
simulation conditions.Besides, a stable trajectory indicated strong binding between them.Further, to identify the flexible and rigid regions of the complex, RMSF analysis was implemented and the average atomic flexibility of the Caatoms of native CDK2 and H2-docked complex were measured.The average RMSF of native (�0.17 nm) and H2 complex (0.18 nm) was relatively higher (Figure 5b).The complex formed displays stable degree of flexibility, exhibiting stable active site residue interactions and at the H-binding residues.

Rg and SASA examination.
The compactness of the receptor-docked complexes was evaluated by radius of gyration (Rg) calculation (Figure 6a).We noted that the native protein showed lower average Rg values (1.89 nm).Furthermore, the complex with H2 also remained stable and enhanced the compactness (1.96 nm), which remained throughout the MD simulation frame of 500 ns.Comparative Rg results revealed stable folding behavior of CDK2 and H2 complex, indicating high compactness of the system under consideration.This suggests that the H2 compounds stayed firmly bound to the active site and helped maintain the stability and compactness of the protein structure.
We also performed SASA analysis to understand the solvent behavior of the complex (Figure 6b).The results revealed that the native and docked complex had very close average values (167.24 and 167.58 nm 2 , respectively), indicating hydrophobic contacts between the receptor and ligand in the complex.

Correlation between drug-likeness and in vitro activity
The fact that drug-likeness studies give a good idea about a prospective drug candidate, attempts have been made to determine drug-likeness features of the reported compounds and correlate with their in vitro activities.Generally, a compound is likely to behave as a drug for clinical uses if it obeys 'Lipinski's rules of Five (Ro5)' (Lipinski, 2000(Lipinski, , 2004;;Lipinski et al., 2012).According to the rule, compounds should have logP (octanol-water partition coefficient) � 5, MW (molecular weight) � 500, HBA (hydrogen bond acceptor) < 10, and HBD (hydrogen bond donor) < 5 to behave as an ideal drug candidate.In addition to the parameters mentioned above, factors like PSA (polar surface area) � 140 � Å 2 , RB � 13 RBs (numbers of rotatable bonds), etc., also play a significant role in drug metabolism (Keller et al., 2006;Veber et al., 2002).It has been suggested that a compound violating two or more properties might not behave as a drug (Lipinski et al., 2001).The Lipinski's parameters and others in simple colour and shape are depicted in Table 3.As it is clear, intermediate compounds S1 and S2 showed higher drug-likeness, with S1 better than S2.Indeed, the presence of 4-bromophenyl at the 3-position of the pyrazole core is responsible for this (logP ¼ 3.4 vs 4.02).In the case of hybrids, except H3, all showed similar and acceptable properties.Except for the logP values, all other features remain within the limit, suggesting suitability of the reported compounds as drug.PSA and RB, which is correlated to oral bioavailability was also within the cutoff values.Overall, this study indicates that the: a.The intermediate compounds (S1 and S2) have higher drug-likeliness than the final compounds, and b.Among final hybrid compounds, the order of drug-likeness was H1 > H4 � H2 > H3.
Interestingly, this observation is corroborated by the experimental results.As discussed in earlier, compound S1 exhibits the best activity, followed by H1 > H4 > H2 > H3 against A549 cancerous cell line.

Bioavailability and toxicology prediction
To get further insight into the pharmacokinetics, bioavailability and toxicological profile of the reported molecules, ADME (absorption, distribution, metabolism and excretion) properties were determined (Daina et al., 2017).The data were generated using SwissADME web tool, and results are gathered in Table T1 (supporting information) and Figure 7a, b.An advantage of this technique is that it include additional factors such as instauration, flexibility, etc., and the output is depicted in the form of radar plot and boiled egg diagram.A molecule falling within the radar plot's pink area can be considered drug-like, while any off-shoot from this plot tends it off from the drug-likeness.Compounds S1 and S2 showed only one off-shoot (insaturation) while H1-H4 showed two or more (lipophilicity, insolubility, or size).Compared to others, H1 showed the least deviation, while H3 showed the most.Overall, this study also indicates the suboptimal physicochemical properties and oral bioavailability of the compounds.Regarding the possibility of passing the blood-brain barrier (BBB), molecules falling in the egg's yolk might surpass this barrier.Compounds S1 and S2 are more likely to pass BBB than the hybrid compounds.Among final compound, the order was: H1 > H4 � H2 > H3 (Figure 7a).
In addition, we also predicted the toxicity profiles of the compounds using the web-based platform pkCSM, and the results are given in Table 4 (Pires et al., 2015).Results of this study indicated that the reported compounds do not bear AMES toxicity (mutagenicity), skin sensitivity or hERG I inhibitory activity.However, others might induce hERG II inhibition and hepatotoxicity (except S2 and H3).The estimated maximum tolerated dose for human use was 0.73-0.81log mg/kg/day, which is relatively better than the structurally similar compounds.However, further in vivo studies are required to confirm these calculations.

Figure 1 .
Figure1.Chemical structures of some pyrazole-chalcone hybrids with anticancer effect on human cancer cell lines reported previously (i-iv) and in this work (v).
, supporting information). 1H-NMR of P1 and P2 showed signals at d 2.41-2.28ppm and d 7.92-7.30ppm attributed to methyl (CH 3 ) and aromatic protons.Hydrazone (-C ¼ N) formation was confirmed by the absence of carbonyl carbon of ketone (C ¼ O) and the presence of C ¼ N linkage at � d144.7-145.9ppm.POCl 3 /DMF mediated cyclization of P1 and P2 yielded pyrazole aldehydes S1 and S2 were confirmed by the presence of two singlets related to pyrazolic proton (H-attached to C-5) at d 9.01-9.05ppm and aldehydic (-CH) � d10.0 ppm signals in the 1 H-NMR spectra in addition to signals due to aromatic protons.In 13 C-NMR, carbonyl carbon resonated at d185.1-185.2ppm. 1 H NMR spectra of the final pyrazole-chalcone hybrids H1-H4 showed singlet at d 9.08-d 9.11 ppm due to one pyrazolic proton (H-attached to C-5) and a doublet at d 8.11 À 8.05 due to the two alkenic protons. 13C-NMR of the compounds showed signals due to enone-fragment of the molecule, that is, carbonyl (C ¼ O at d 189.36-187.29 ppm) and alkene (C ¼ C at d 121.6-122.32ppm) fragments.

Figure 2 .
Figure 2. Cytotoxicity assay of the compounds against human umbilical vein endothelial cells (HUVEC).

Figure 4 .
Figure 4. Molecular docking results of the studied compounds.CDK2 surface model (top) and residues involved in H-bond formation (bottom) are shown only.

Figure 3 .
Figure 3. MTT anticancer assay of the compounds against (a) human lung (A549) and (b) colorectal (Caco-2) adenocarcinoma cell lines.(c) Doxorubicin was used as a positive control.

Figure 6 .
Figure 6.Radius of gyration (Rg) and SASA during MD simulations (a) Rg plot and (b) SASA analysis plot of 2VTO-docked complexes of the native protein (black colour) and H2-complex (red color).

Table 1 .
Binding energy and interacting residues of compounds.

Table 2 .
RMSD, RMSF, Rg, and SASA values for the native protein and H2-complex.

Table 4 .
Toxicity prediction results of the selected compounds.