N-phenyl-6-chloro-4-hydroxy-2-quinolone-3-carboxamides: Molecular Docking, Synthesis, and Biological Investigation as Anticancer Agents

Cancer is a multifactorial disease and the second leading cause of death worldwide. Diverse factors induce carcinogenesis, such as diet, smoking, radiation, and genetic defects. The phosphatidylinositol 3-kinase (PI3Kα) has emerged as an attractive target for anticancer drug design. Eighteen derivatives of N-phenyl-6-chloro-4-hydroxy-2-quinolone-3-carboxamide were synthesized and characterized using FT-IR, NMR (1H and 13C), and high-resolution mass spectra (HRMS). The series exhibited distinct antiproliferative activity (IC50 µM) against human epithelial colorectal adenocarcinoma (Caco-2) and colon carcinoma (HCT-116) cell lines, respectively: compounds 16 (37.4, 8.9 µM), 18 (50.9, 3.3 µM), 19 (17.0, 5.3 µM), and 21 (18.9, 4.9 µM). The induced-fit docking (IFD) studies against PI3Kαs showed that the derivatives occupy the PI3Kα binding site and engage with key binding residues.


Introduction
The emergence of cancer cases continues to rise worldwide. Cancer, a group of diseases, is considered a prevalent health problem and cause of death. Cancer is uncontrolled cell proliferation accompanied with invasion to other body organs [1]. Multiple factors induce cancer occurrence, such as smoking, diet, radiation, viral infections, and genetic disorders [2][3][4][5]. Phosphatidylinositol 3-kinases (PI3Ks) catalyze the phosphorylation of phosphatidylinositol on the 3-OH group of inositol ring generating phosphatidylinositol 3,4,5 triphosphates (PIP 3 ). PIP 3 invoke downstream effectors and signaling proteins such as protein kinase B (AKT) inducing cell growth and multiplication [6,7].
The phosphatase and tensin homolog (PTEN) dephosphorylates the 3-OH group of PIP 3 producing PIP 2 and therefore antagonizes PI3Ks [7,8]. PI3Ks are organized into three classes (I, II, and III) according to their substrate interaction and protein sequence. Class IA PI3Ks accommodate PI3Kα, β, and δ isoforms encoded by their corresponding genes PIK3CA, PIK3CB, and PIK3CD [9]. Irregular activation of the PI3Kα/AKT signaling pathway has been observed in multiple human cancers [10]. The PIK3CA is catalyzed, over encrypted, and altered in human cancers [10]. Mutations of PI3Kα domains, the helical (E545K and E542K) and kinase (H1047R), are identified in breast, brain, GIT, and uterus cancers [11][12][13][14]. These carcinogenic transformations deform PI3Kα trajectories and thus in turn motivate researchers to design and develop selective PI3Kα inhibitors [13]. The ubiquity of PI3Kα and PTEN alterations in diverse human cancers highlights PI3Kα as an attractive target for anticancer drug design [15,16].
In this respect, we designed and synthesized a series of N-phenyl-6-chloro-4-hydroxy-2-quinolone-3-carboxamides to explore the effect of introducing chloro functionality on the inhibitory activity and to clarify the structure-activity relationship (SAR) of the analogues. Biological testing of the verified compounds and pan PI3K inhibitor (LY294002) was probed in vitro against human epithelial colorectal adenocarcinoma (Caco-2) and human colon carcinoma (HCT-116) cell lines.

Chemistry
Verified compounds (7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24) were designed to investigate the effect of attaching different motifs at the carboxamide link of 6-chloro-4-hydroxy-quinolinone core structure and their inhibitory activity. Reaction of 3 and 4 in sodium ethoxide under reflux produced the core nucleus 5 with 88% yield, as represented in Scheme 1. Thin-layer chromatography (TLC) application was recruited to monitor the reaction progress; the absence of a spot belonging to 3 highlighted the completeness of the reaction. Target compounds 7-27 have been generated by refluxing 5 with an excess of aromatic amine (ArNH 2 ) using DMF (dimethylformamide.), as depicted in Table 1 and Scheme 1. The disappearance of a spot of compound 5 on the TLC plate suggested that the reaction was finished. The identity of the chemical backbones was confirmed using IR, NMR and high-resolution mass spectra (HRMS). The elaborate experimental and spectroscopic data agree with the prospective structures.
Screening against Caco-2 showed that compounds tailored with o-CF 3 (8) or m-CF 3 (9) provoke the activity suggesting that hydrophobic and/or H-bond interaction mediate(s) ligand/PI3Kα complex formation. Similarly, the antiproliferative activity of p-CF 3 (10) implies that hydrophobic and/or H-bond interaction guides (s) ligand/PI3Kα interaction at p-position.
The antiproliferative activity of p-OCH 3 (11) and p-CH 3 (12) proclaims that oxygen atom places the -CH 3 deeply in the binding pocket close to the hydrophobic lining, thus inducing the activity. Additionally, the activity of p-CF 3 (10) and p-CH 3 (12) indicates that H-bond drives ligand/PI3Kα interaction on p-position. Interestingly, elongation of the carboxamide side chain by one carbon exemplified by 7 enhances the activity assuming that the aromatic motif is buried deeply in the hydrophobic site. This result presumes that elongation of the carboxamide side chain (7) or tailoring the phenyl by -OCH 3 (11) situates the hydrophobic motif in the hydrophobic cleft. Tailoring the carboxamide side chain with substituted benzoic acid moiety (13, 14, and 15) reveals that H-bond mediates ligand/PI3Kα binding on o-and m-positions and thus in turn provides an explanation for the activity of 8 and 9. Furthermore, the activity of 14 suggests that the carboxylic group on o-position might be involved in intra H-bond generating a conformer that better orientates the analogue in the binding domain. Also, the activity of 13, 14, and 15 anticipates the substantial effect of choloro motif on better accommodation in the binding cleft. Introducing a heterocyclic group on the side chain of carboxamide illustrated by (16 and 17) confirms the significance of the aromatic or heterocyclic group for better activity. Additionally, the nitrogen atom in pyridine provides H-bond acceptor and promotes water solubility and polarity.
The activity of 18 affirms the importance of substituted benzoic acid motif and declares that the ionized carboxylate anion might impede cell membrane permeability. Additionally, the activity of 18 provides a clue for the beneficial influence of choloro group on cell membrane permeability as exemplified by 13, 14, and 15. Attaching a fluorinated benzene moiety on carboxamide side chain (19, 20, and 21) induces the activity interrogating that the fluoro moiety on o-, m-, and p-positions properly accommodates the binding site due to its small size. Introducing the methylated benzoic acid motif (22 and 23) elicits the activity inferring that hydrophobic interaction mediates ligand/PI3Kα interaction on o-and m-positions corresponding to -COOH. Contrarily, methoxylated benzoic acid moiety (24) induces the activity implying that H-bond and/or hydrophobic interaction on p-position, relatively to-COOH, drives ligand/PI3Kα binding.
Screening against HCT-116 revealed that elongating the carboxamide side chain (7) induces the inhibitory activity inferring better accommodating in the binding domain. The biological data of o-CF 3 (8), m-CF 3 (9), and p-CF 3 (10) declare that hydrophobic and/or Hbond mediates ligand/PI3Kα interaction on o-position. The inhibitory activity of p-OCH 3 (11) anticipates that the oxygen atom might push the -CH 3 properly in the binding domain. Furthermore, the activity of p-CH 3 (12) confirms the significance of oxygen atom and provides further explanation for the activity of p-OCH 3 (11). Additionally, the inhibitory activity of p-CF 3 (10) and p-CH 3 (12) elucidates the importance of H-bonding at the pposition. Attaching a substituted benzoic acid moiety on carboxamide side chain (13, 14, and 15) shows that the p-chloro moiety provokes the activity anticipating a favored conformation is mediated by intra-H-bond between -COOH and the carboxamide motif. Tailoring the carboxamide side chain with pyridine-4-yl (16) induces the activity inferring that H-bond and/or ionic interaction mediates 16/PI3Kα complex formation. Additionally, the activity of pyridine-3-yl (17) affirms the significance of pyridine-4-yl on inducing the inhibitory activity.
Attaching 3-benzoic acid moiety (18) elicits the activity suggesting that H-bond and/or ionic interaction mediates ligand/PI3Kα interaction. Incorporating a fluorinated benzene moiety (19, 20, and 21) shows that the o-and p-fluoro analogs mediate H-bond with key binding residues. Attaching a methylated benzoic acid motif (22 and 23) suggests that the intra-H bond offered by -COOH and the carboxamide motif generates a conformer that impedes their proper orientation in the binding domain. The activity of methoxylated benzoic acid moiety (24) declares that the oxygen atom might push -CH 3 in the binding sites and the existence of -COOH might provide an intra-H-bond that orientates 24 properly in the binding site. Furthermore, the difference in the biological activity of 16, 18, 19, 21, and 24 against HCT-116 proclaims that selectivity is attained. Finally, the activity of 5 against Caco-2 and HCT-116 cell lines emphasizes the significance of substituted carboxamide motif.
Noting that the probed cell lines express both WT and MUT PI3Kαs, separating either enzyme is strongly advised to validate the suppressive activity against isolated PI3Kα. Future confirmation should be accomplished by silencing either gene to explore the antiproliferative mechanism against each PI3Kα.
The effects of compounds (18 and 19) on the PI3K/AKT signaling pathway were investigated using qPCR. The relative gene expression of PI3K, AKT, and BAD (BCL2 Associated Agonist of Cell Death) was significantly changed by treatment with either compound (0.5 µM) in a manner consistent with the positive control treatment (0.5 µM) ( Figure 3). A substantial reduction in PI3K and AKT gene expression was observed upon treatment with 18, 19, or the positive control LY-294002. On the other hand, the expression of the pro-apoptotic gene BAD was significantly increased in treated cells (for all tested compound including the positive control) when compared to the negative control where there was no treatment applied.
Remarkably, 5, 7-24 showed an affinity against both WT and MUT (H1047R) PI3Kα. Fortunately, higher binding affinity against 3HHM was recorded for 12 and 15. Around 2 Kcal difference in binding energy against PI3Kαs might predict that this series is mutant selective. Furthermore, the dominance of S774 and K802 in ligand/PI3Kα complex formation infers that the scaffold might be PI3Kα-selective inhibitors [19].
To evaluate the execution of the IFD algorithm, we compared the docked pose of X6K in WT PI3Kα (PDB ID: 4L23) [39] to its native geometry in the crystal assembly. Figure 6 demonstrates the overlaying of the IFD-extracted X6K conformation and its native geometry in 4L23. The RMSD for heavy atoms of X6K between IFD-extracted docked conformation and the native pose was 0.169 Å. The result infers that IFD can determine the native geometry in crystal co-ordinates and prosperously anticipate the ligand binding pose.  In order to inspect the binding groups of 7-24, we screened them against a reported pharmacophore model of active PI3Kα inhibitors [18]. The core structure of 7-24 fit the functionalities of active PI3Kα inhibitors (Figure 7

Descriptor Analysis
Sixty-seven molecular descriptors comprising two main categories of alvaDecs [41] (i.e., drug-like indices, molecular properties and P_VSA-like descriptors) were calculated for molecules Mol.5-Mol.24. All descriptor values were then analyzed using principal component analysis. Our principal component analysis reveals that our compounds clustered into three main groups in the 3-D space formed by the first three principle components for the descriptor variance ( Figure 8A). All calculated descriptors for all analyzed molecules are reported in Supplementary Table S1. Our analysis further revealed that our synthesized molecules differ in their lead-like and some drug-like properties as shown in Figure 8B.

Chemistry
Chemical reagents and solvents were of a highly purified grade and used right away lacking any purification technique. Reagents were bought from Alpha-Chemika (Mumbai, India), Acros Organics (Fair Lawn, NJ, USA), Sigma-Aldrich (St. Louis, MI, USA), and Scharlau Company (Barcelona, Spain). Rota vapor (1) model R-215 (Buchi, Switzerland) attached to vacuum pump model v-700 and water bath B-491 were recruited to evaporate ordinary solvents and vacuum controller v-855 was used to evaporate high boiling solvents such as DMSO and DMF.
Melting points (MP) were recorded by Gallenkamp melting point apparatus. Hot plate and magnetic stirrer were delivered by vision scientific CO, LTD USA. Thin-layer chromatography (TLC) was conducted on 20 × 20 cm 2 with layer thickness 0.2 mm aluminum cards pre-coated with fluorescent silica gel GF254 DC (Fluka analytical, Buchs, Switzerland) and detected by UV light indicator (at 254 and/or 360 nm).
Infrared (IR) spectra were recorded as, KBr discs, with a Shimadzu IR Affinity FTIR spectrophotometer. Nuclear magnetic resonance (NMR) 1 H-and 13 C-NMR spectra were measured on Bruker NanoBay 400 MHz spectrophotometer (the Hashemite University, Zarqa, Jordan). Chemical shifts were stated in δ (ppm) using TMS as an internal reference. High-resolution mass spectra (HRMS) were measured using a Bruker APEX-IV (7 Tesla) instrument. External calibration was performed using an arginine cluster at a mass range of m/z 175-871.

MTT Assay
All cells were plated at a density of 8 × 10 3 cells per well in 96-well plates and incubated to allow attachment for 24 h. The in vitro evaluation of the antiproliferative activities of the examined series was accomplished using the 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl tetrazolium bromide (MTT) colorimetric assay, as previously described [42]. In short, compounds were diluted in culture media to yield the required concentration and applied to test wells for 48 h at 37 • C in a 5% CO 2 incubator. Three triplicates of each concentration for all tested compounds were conducted in three independent assays (n = 9). DMEM samples were employed as negative controls, and LY294002 as a positive control. At the end of the exposure period, 20 µL of 0.5 mg/mL of MTT was added to each well and incubated for 4 h, afterword its reduction to formazan by metabolically active cells was calculated via measuring the absorbance at 570 nm. Cell viability was calculated based on the measured absorbance relative to the absorbance of the cells exposed to the negative control, which represented 100% cell viability.

Statistical Analysis
Data analysis was performed using GraphPad Prism software version 7. The differences between treatments groups were determined by t-test or one-way analysis of variance (ANOVA) followed by Tukey post hoc t-test as appropriate. Data were expressed as mean ± SD and p < 0.05 was considered a statistically significant difference. A non-linear regression analysis was used to calculate IC 50 values.

Quantitative Real-Time PCR RNA Extraction
Total RNA was extracted from the cultured cells using Quick-RNA MiniPrep according to the manufacturer's instructions as follows: harvested cells were re-suspended in RNA lysis buffer and centrifuged, the supernatant then was transferred into a Spin-Away Filter in a collection tube and centrifuged, and then 95% ethanol was added and mixed well. The mixture was then transferred to a Zymo-Spin IIICG Column in a collection tube and centrifuged, followed by adding RNA Prep Buffer to the column, followed by two washing steps by adding RNA wash buffer to the column and then centrifugation. Finally, the RNA was eluted by adding DNase/RNase-free water to the column and centrifuged. Samples were quantified using a spectrophotometer via absorbance at 260/280 nm.

Complementary DNA (cDNA) Synthesis
Complementary cDNA synthesis was performed using a ProtoScript First Strand cDNA Synthesis Kit following the manufacturer's instructions. In brief, 1000 nanogram from total RNA was add to a total 20 µL reaction volume that included 60 µM random primer mix, reaction mix, 10 × enzyme mix, and nuclease-free water. The tubes were then incubated at 25 • C for 5 min followed by incubation at 42 • C for 1 h. The cDNA was then then stored at −20 • C until further analysis.

Real-Time PCR
The sequences of the primers that were used in real-time PCR assay are shown in Table 4. The reaction mixtures consisted of 200 ng cDNA template, 10 µM of each primer, 10 µL of 2X SYBER Premix Ex Taq II (Takara BIO INC, Shiga, Japan), and the total reaction volume was 20 µL. The reaction was carried out using a BIO RAD iQ5 Multicolor real-time PCR Detection System thermal cycler under the following reaction conditions: 1 cycle of 2 min at 95.0 • C, followed by 45 cycles of 10 s at 95.0 • C, 25 s at 57.0 • C, 25 s at 60.0 • C, and a final cycle of 30 s at 55.0 • C. To confirm that only one PCR product was amplified, dissociation curve analysis of amplification products was performed at the end of the last amplification cycle. The co-ordinates of WT PI3Kαs (PDB ID: 2RD0) [9] and (PDB ID: 4L23) [39] as well as MUT (H1047R) PI3Kα (PDB ID: 3HHM) [35] were obtained from the RCSB Protein Data Bank . The template of X6K in 4L23 [39] was conveyed to 2RD0 and pinpointed as a ligand to extract the grid file. The homology modeled assemblies of 2RD0 and 3HHM were previously reported by our research group [19] and employed for this study. The co-ordinates of 2RD0 and 3HHM were energetically minimized to avoid steric clash. Extra treatment for the minimized co-ordinates was executed using Protein Preparation algorithm in Schrödinger software [38] to maximize H-bond interactions between side chains.

Preparation of Ligand Structures
The verified compounds (ligands) were modeled using wortmannin's template in 3HHM. The ligands were built by MAESTRO [38] Build script and energetically minimized by the MacroModel algorithm using OPLS2005 force field.

Induced-Fit Docking (IFD)
The adopted ligand (wortmannin) was labeled as a centroid in 2RD0 [9] and 3HMM [35] kinase domains. The Van der Waals scaling factors for receptor and ligand were calibrated to 0.5 to provide adequate flexibility for the best docked ligand pose. Other parameters were adjusted as default. The ligand geometry with the highest XP Glide binding affinity was recorded.

Molecular Descriptors
All molecular structures, sketched in ChemDraw [46] and saved in SDF file format, were standardized according to the methods described by Hajjo et al. [47]. Next, two groups of molecular descriptors, comprising 'Drug-like Indices' and 'Molecular Properties' calculated using alvaDesc software from Kode Cheminformatics [48], were generated for compounds, 5,7-27.

Principal Component Analysis (PCA)
A principal component analysis was performed on structures 5 and 7-27 using druglike indices and molecular properties. All calculations and PCA analysis were performed using alvaDesc software from Kode Cheminformatics [48].
Biological investigation in Caco-2 and HCT116 cell lines showed potent suppression HCT-116 mediated by compounds (16, 18, 19, 21, and 24). IFD studies against PI3Kαs illustrated that the series accommodate PI3Kαs binding sites and engage with key residues involved in inhibitor interactions. We are looking for optimizing the core structure of this series to induce its anticancer activity and selectivity against a panel of kinases.