Synthesis of 9-Hydroxy-1H-Benzo[f]chromene Derivatives with Effective Cytotoxic Activity on MCF7/ADR, P-Glycoprotein Inhibitors, Cell Cycle Arrest and Apoptosis Effects

β-Enaminonitriles bearing 9-hydroxy-1H-benzo[f]chromene moiety was synthesized. The targeted compounds were evaluated for their anti-proliferative activity against three human tumor cell lines, PC-3, SKOV-3 and HeLa, and the active cytotoxic compounds were further evaluated against cancer cells, MCF-7/ADR, and two normal cell lines, HFL-1 and WI-38. Few compounds were assigned to be the most potent derivatives against PC-3, SKOV-3 and HeLa cell lines in comparison with Vinblastine and Doxorubicin. Several compounds possessed a relatively good potency against MCF-7/ADR cells as compared with Doxorubicin and were tested as a P-gp inhibitor. Moreover, the halogenated substituents, 2,4-F2, 2,3-Cl2, 2,5-Cl2 and 3,4-Cl2; have good potency against P-gp-mediated MDR in MCF-7/ADR as compared with Doxorubicin. Meanwhile, Rho123 accumulation assays revealed that few compounds effectively inhibited P-pg and efflux function. In addition, certain derivatives induced apoptosis and an accumulation of the treated MCF-7/ADR cells in the G1, S and G1/S phases.

Overexpression of ABC transporters such as P-glycoprotein (P-gp), multidrug resistance protein 1 (MDR1) and ATP-binding cassette subfamily B member 1 (ABCB1) is the main contributor to MDR [26,27]. On the other hand, MDR is contributing to chemotherapy failure [28]. The multidrug-resistant phenotype in cancer that mediates MDR also heavily relies on P-glycoprotein (P-gp/ABCB1) [29].
Overexpression of ABC transporters such as P-glycoprotein (P-gp), multidrug resistance protein 1 (MDR1) and ATP-binding cassette subfamily B member 1 (ABCB1) is the main contributor to MDR [26,27]. On the other hand, MDR is contributing to chemotherapy failure [28]. The multidrug-resistant phenotype in cancer that mediates MDR also heavily relies on P-glycoprotein (P-gp/ABCB1) [29]. Based on our ongoing experiences to create effective anticancer benzochromenebased medicines, [30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48], we have initiated the current study. Developing drugs with potential anti-proliferative activity against three cancer cell lines, PC-3, SKOV-3 and HeLa, while having no cytotoxic effect on normal cell lines HFL-1 and WI-38, was the main motivation behind our decision to design and synthesize β-enaminonitriles incorporating a 9-hydroxy-1H-benzo[f]chromene moiety. An additional objective is to target MCF-7/ADR These compounds were found to be effective against P-gp inhibitors. Additionally, compounds 4d and 4e caused apoptosis and an increase in MCF-7/ADR cells in the G1 phase, whereas compounds 4a, 4c, 4g and 4i did so in the S phase and G1/S phases, respectively ( Figure 1). The type of substituents at the 9-position and the aryl group at the 1-position appears to be critical to the cytotoxic action; therefore, this was well thought out during the rational design of the 9-hydroxy-1H-benzo[f ]chromene derivatives. Figure 2 compares the behaviors of the recently synthesized derivatives linked to a hydroxyl group at position 9 with those previously synthesized with unsubstituted naphthalene moiety: in first generation 1 and 2 [49] is a substituted naphthalene moiety linked to a bromine at position 9, second generation is 3-7 [10] and third generation with a methoxy group at position 9 is 8-12 [50]. The comparison revealed that the newly synthesized derivatives 4a, 4c, 4d, 4e, 4g, 4i, 4k and 4o significantly increased or paralleled the potency of the previously synthesized compounds 1-12 when used against cancer cell lines [10,49,50].  Figure 1). These compounds were found to be effective against P-gp inhibitors. Additionally, compounds 4d and 4e caused apoptosis and an increase in MCF-7/ADR cells in the G1 phase, whereas compounds 4a, 4c, 4g and 4i did so in the S phase and G1/S phases, respectively ( Figure 1). The type of substituents at the 9-position and the aryl group at the 1-position appears to be critical to the cytotoxic action; therefore, this was well thought out during the rational design of the 9-hydroxy-1H-benzo[f]chromene derivatives. Figure   Figure 2. Rationale for designing target benzochromene derivatives and previous studies compounds.

Optical Activities
A Carl Zeiss polarimeter (Jena, Germany.) was used to gauge the target compound 4 which found "optical inactive" in the form of a racemic (±) mixture [51], refer to Scheme 1.

Spectroscopic Data
The structures and purities of the synthesized compounds 4a-p were confirmed by spectral analysis, i.e., IR, 1 H NMR, 13 C NMR and MS spectra. The IR spectra of 4a-p revealed the presence of peaks around υ 3490-3423, 3445-3408, 3353-3222, 3267-3177 cm − assigned to the amino and hydroxyl groups and for the CN groups in the region υ 2214-2175 cm. −1 The 1H NMR spectra of 4a-p showed the singlet signal of the OH, NH2 and CH protons around δ 10.77-9.82, 7.27-6.41 and δ 5.97-4.96 ppm, respectively. Furthermore the 13 C NMR spectra of 4a-p showed the signals resonating at δ 38.47-26.51 ppm, attributable to the CH carbons. In addition, the MS spectra and the elemental analyses of 4a-p delivered absolute confirmation for their structures (see Supplementary Materials of the spectral data from Figure S1 to S55). Scheme 1. Microwave irradiation synthesis of 3-amino-1-aryl-9-hydroxy-1H-benzo[f]chromene-2carbonitrile (4a-p).

Optical Activities
A Carl Zeiss polarimeter (Jena, Germany.) was used to gauge the target compound 4 which found "optical inactive" in the form of a racemic (±) mixture [51], refer to Scheme 1.

Spectroscopic Data
The structures and purities of the synthesized compounds 4a-p were confirmed by spectral analysis, i.e., IR, 1 H NMR, 13 C NMR and MS spectra. The IR spectra of 4a-p revealed the presence of peaks around υ 3490-3423, 3445-3408, 3353-3222, 3267-3177 cm −1 assigned to the amino and hydroxyl groups and for the CN groups in the region υ 2214-2175 cm. −1 The 1H NMR spectra of 4a-p showed the singlet signal of the OH, NH 2 and CH protons around δ 10.77-9.82, 7.27-6.41 and δ 5.97-4.96 ppm, respectively. Furthermore, the 13 C NMR spectra of 4a-p showed the signals resonating at δ 38.47-26.51 ppm, attributable to the CH carbons. In addition, the MS spectra and the elemental analyses of 4a-p delivered absolute confirmation for their structures (see Supplementary Materials of the spectral data from Figure S1 to S55).

In Vitro Cytotoxic Activity
Three human cancer cell lines, PC-3 (human prostate adenocarcinoma, metastatic), SKOV-3 (ovarian cancer cell line) and HeLa (cervical cancer cell line), were used to test the anti-proliferative activity of β-enaminonitrile (4a-p) using the MTT colorimetric assay [52] at concentrations ranging from 0-100 µM. The representative active cytotoxic substances 4a, 4c, 4d, 4e, 4g, 4i, 4k and 4o were also tested against two normal cell lines, human fetal lung (HFL-1) and human diploid fibroblasts (WI-38), as well as Adriamycin (ADR)-resistant human breast cancer cells (MCF-7/ADR). The reference cytotoxic agents, Vinblastine and Doxorubicin, were used in the tests. The results were expressed as growth inhibitory concentration (IC 50 µM) values, which reveal the concentrations necessary for 50% inhibition of cell growth after 24 h of incubation compared to the untreated ones (Figures 3 and 4 and  Table 1).

In Vitro Cytotoxic Activity
Three human cancer cell lines, PC-3 (human prostate adenocarcinoma, metastatic), SKOV-3 (ovarian cancer cell line) and HeLa (cervical cancer cell line), were used to test the anti-proliferative activity of β-enaminonitrile (4a-p) using the MTT colorimetric assay [52] at concentrations ranging from 0-100 μM. The representative active cytotoxic substances 4a, 4c, 4d, 4e, 4g, 4i, 4k and 4o were also tested against two normal cell lines, human fetal lung (HFL-1) and human diploid fibroblasts (WI-38), as well as Adriamycin (ADR)-resistant human breast cancer cells (MCF-7/ADR). The reference cytotoxic agents, Vinblastine and Doxorubicin, were used in the tests. The results were expressed as growth inhibitory concentration (IC50 μM) values, which reveal the concentrations necessary for 50% inhibition of cell growth after 24 h of incubation compared to the untreated ones ( Figures 3 and 4 and Table 1).

P-Glycoprotein Expression and Function in MCF-7/ADR Cell Lysate
P-glycoprotein (P-gp)-mediated multidrug resistance (MDR) is a phenomenon in which cells become resistant to structurally and mechanistically unrelated drugs, resulting in low intracellular drug concentrations. It is one of the noteworthy problems in malignant tumor clinical therapeutics [53]. The most active compounds 4a, 4c-4e, 4g and 4i against MCF-7/ADR cells were tested as P-gp inhibitors using commercial ELISA kits (Human permeability glycoprotein ELISA Kit). The inhibition of P-gp content in MCF-7/ADR cell lysate using different compounds conc. as shown in Figure 5 and Table 2. In the near future, we will assess this in an in vivo model using MDA-MB-231 xenografts grown on the chorioallantoic membrane of fertilized chick eggs (CAM). Additionally, we will then study the drug likeness for the benzochromene derivatives.

P-Glycoprotein Expression and Function in MCF-7/ADR Cell Lysate
P-glycoprotein (P-gp)-mediated multidrug resistance (MDR) is a phenomenon in which cells become resistant to structurally and mechanistically unrelated drugs, resulting in low intracellular drug concentrations. It is one of the noteworthy problems in malignant tumor clinical therapeutics [53]. The most active compounds 4a, 4c-4e, 4g and 4i against MCF-7/ADR cells were tested as P-gp inhibitors using commercial ELISA kits (Human permeability glycoprotein ELISA Kit). The inhibition of P-gp content in MCF-7/ADR cell lysate using different compounds conc. as shown in Figure 5 and Table 2.    Table 2 indicated that compounds 4a, 4c, 4e and 4g that have 2,4-F 2 , 2,3-Cl 2 , 2,5-Cl 2 and 3,4-Cl 2 substituents have good potency against P-gp in MCF-7/ADR with IC 50 ranging from 18.7 to 34.6 µM as compared with Doxorubicin (IC 50 = 50.9 µM), as shown in Figure 5 and Table 2. Despite having a high cytotoxic effect on MCF-7/ADR cells, chemicals 4d and 4i had an almost negligible inhibitory effect on the P-gp level ( Figure 5); however, only 4a, 4c and 4e had the strong inhibitory effect needed to be highly effective in reversing the MDR in MCF-7/ADR in earlier studies [54,55]. Furthermore, the reversion of P-gp-mediated multidrug resistance could be achieved by the down-regulation of P-gp expression and/or P-gp efflux function inhibition [56]. Hence, the effect of the current synthesized drugs was tested for inhibitory potential of P-glycoprotein activity using a Rhodamine 123 Accumulation Assay (Rhodamine Competitive ELISA Kit).
The information presented in Table 3 shows that Rhodamine 123 has a (IC 50 ) range of 14.6-28.7 µM for evaluating the P-gp functional inhibition of 4a, 4c and 4e in comparison with 14.3 µM referenced medication "Verapamil". The compounds 4a, 4c and 4e reduced P-gp expression and its function as well, which had an impact on the recovery of sensitivity to MCF-7/ADR cells. Contrarily, the cytotoxic effect of 4d, 4g and 4i against MDR could be attributed to genetics, growth hormones and/or improved DNA repair capability [57].

Cell Cycle Arrest in Treated MCF-7/ADR Cancer Cells
Cancer cells undergo unscheduled cell divisions by the down-regulation of the four cell cycle stages (G1, S, G2 and M); therefore, the development of anticancer agents targeting cell cycle arrest represents an important therapeutic intervention [58,59]. P-gp breast cancer resistance proteins (BCRP) inhibit the action of anticancer drugs on cell cycle arrest by preventing intercellular drug accumulation [60]. By using flow cytometry and the FACS Calibers, it was possible to examine how the potent-produced compounds (4a, 4c, 4e, 4g and 4i) affected the regulation of cell cycle progression in MCF-7/ADR cells (Becton Dickinson). The distribution of cells along the G1 (2n), G2/M (4n) and S (2n to 4n) phases of the cycle was exhibited in the representative cell cycle distribution histogram of the stained DNA (Figure 6a). The MCF-7/ADR cancer cells were remedied with each derivative at its IC 50 values for 24 h, a controlled experiment with no treatment given with " blank trials".
The outcomes of the cell cycle progression give three different results showing compounds with cell cyclic arrest at the G1 phase (4d and 4e), while there were other compounds at the S phase (4a and 4c) and a final compound with cell cycle arrest at both of the G1/S phases (4g and 4i). Additionally, in comparison to the untreated controlled cells, these results were followed by a significant drop in percentage at the G2/M stages (Figure 6b). According to the cell cycle analysis, the tested derivatives considerably stopped the MCF-7/ADR cancer cells in the G1/S phase.

Apoptosis Induction in MCF-7/ADR Cancer Cells
One of the optional mechanisms regarding the lethal action of chemotherapeutic medicines is blocking cell cycle progression, triggering apoptosis with otherwise mutual effect [61]. Moreover, P-gp prevents the release of cytochrome c, which is controlled by the intrinsic mitochondrial pathway, hence inhibiting apoptosis [62]. The pivotal relation of the newly synthesized MCF-7/ADR "anticancer drugs" was further evaluated using the Annexin V/PI double staining flow cytometric test [63]. Figure 7a shows specimen dot plots of the double-stained MCF-7/ADR cells following treatment with the various substances under investigation. In comparison to Doxorubicin's 30% total apoptosis rate, which was nearly identical to 4a, 4c and 4g, compounds 4d, 4e and 4i were leading with percentages of 41%, 31% and 38%, respectively. Only compounds 4a (8.5%) and 4g (11%) had greater effects on necrosis compared to doxorubicin (5%). Other examined substances caused both early and late apoptosis in all treated cells (Figure 7b). Therefore, MCF-7/ADR cytotoxicity is induced by apoptotic processes, apparently without negative effects on P-gp.

Molecular Docking
In order to confirm the link between the in vitro cytotoxicity results and the binding affinity of hybrids 4a-p, docking simulation experiments were carried out. More attention is devoted to P-glycoprotein (P-gp) which exists in cryo-EM structure; it contains the active site (PDB ID: 6FN1 [64]). The active site for 6FN1 was determined using CASTp to be Trp231, 1GLU874, MET875, GLN945 and MET948. The 3D loop structure of 6FN1 was created by the mGenTHERADER [65] within the docking framework, while the TS protein was created using the PyMol v:1.4 package. Nevertheless, the P-gp protein was validated using PROCHECK server [66], which also provided a 97.36 overall quality factor score on ERRAT server, to overview the statistics of nonbonding interactions between various atom types. When compared to the reference drug Zosuquidar, which blocked Glu874 and Trp231 with an RMSD of 1.20 Å, the toxicity behavior of 4a-p molecules was demonstrated in binding energy (BE) with a P-gp receptor. After being re-docked, the studied 4a-p hybrids attained a root mean square deviation (RMSD) of less than 2 Å (Table 4). To evaluate the binding affinity of the synthesized compounds, the pose with the lowest BE and RMSD was chosen. To further support the validity of the BE, the inhibitory constant (Ki) and ligand efficacy (LE) were calculated [67]. The most active compounds (4a, 4c, 4d, 4e and 4i) display promising binding scores (−6.30, −7.17, −7.10, −8.12 and −6.07 Kcal/mol), respectively, compared to Zosuquidar (−8.22 Kcal/mol). These BE values are supportive to the cytotoxicity data for tested compounds. All compounds were successful in blocking active binding sites in a manner similar to Zosuquidar. The docking statistical analysis implemented herein favors 30% of the tested compounds and can inhibit the subpocket P-gp protein by blocking the vital amino acid Glu 874 in the binding pocket (Figure 8). metrics LE, Ki and FQ were found within the traditional range [68]. As per the docking outcomes, P-gp protein inhibition supports the in vitro findings. Zosuquidar 4a 4c -4d 4e 4i PLIF statistical analysis Figure 8. Binding interaction of most active compounds 4a, 4c, 4d, 4e and 4i into P-gp, and PLIF statistical analysis histogram for compounds 4a-p and residues of P-gp protein. The most active compound 4e bearing 2,5-Cl 2 substituents (IC 50 18.7 µM) showed the most binding affinity −8.12 Kcal/mol and inhibited the binding site by the formation of two strong H-bonding interactions with Tyr309, Gln7724 and Phe342 (Figure 8). The compound showed promising bioactive parameters LE, Ki and FQ (Table 4) but moderate activity was obtained for compounds 4a and 4c bearing 2,4-F 2 and 2,3-Cl 2 substituents.

Materials and Equipment
The 2,4-F 2 and 2,3-Cl 2 rings are arranged with the main GLU874 and TRP231 in a perpendicular manner to be stabilized in the binding pocket, in addition to strong Hbonding interactions with the respective amino acids. For all compounds, the bioactivity metrics LE, Ki and FQ were found within the traditional range [68]. As per the docking outcomes, P-gp protein inhibition supports the in vitro findings.

Materials and Equipment
All chemicals were purchased from Sigma-Aldrich Chemical Co. (Sigma-Aldrich Corp., St. Louis, MO, USA). All the melting points were measured with a Stuart Scientific Co. Ltd. Apparatus (Camberley, UK), which means they are uncorrected. The IR spectra were recorded on a KBr disc on a Jasco FT/IR 460 plus spectrophotometer (Jasco, Tokyo, Japan). The 1 H/ 13 C (500/125 MHz) NMR spectra were measured on a BRUKER AV 500 MHz spectrometer (BRUKER, Billerica, MA, USA) in DMSO-d 6 , as solvent, using tetramethylsilane (TMS) as an internal standard. The microwave apparatus utilized is Milestone Sr1, Microsynth (Milstone, Shelton, CT, USA.). The mass spectra were determined on a Shimadzu GC/MS-QP5050A spectrometer (Shimadzu, Kyoto, Japan). The elemental analysis was carried out at the Regional Center for Mycology and Biotechnology (RCMP), Al-Azhar University, Cairo, Egypt, and the results were within ± 0.25%. The reaction courses and products were routinely monitored using thin layer chromatography (TLC) on silica gel precoated F 254 Merck plates (Merck, Billerica, MA, USA).

General Procedure for Synthesis of 1H-Benzo[f]chromene Derivatives (4a-p)
A reaction mixture of naphthalene-2,7-diol (1) (0.01 mol), malononitrile (2) (0.01 mol), different aromatic aldehydes (3a-p) and piperidine (0.5 mL) in an absolute ethanol solution (30 mL) was heated under microwave irradiation conditions for 2 min at 140 • C. After the reaction reached completion, the reaction mixture was cooled to room temperature and the precipitated solid was filtered off, washed with methanol and recrystallized from ethanol/benzene. The physical and spectral data of compounds (4a-p) are as follows: The tumor cell lines MCF-7, HepG-2, PC-3 MCF-7/ADR, HFL-1 and (WI-38) were obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). The cells were grown on RPMI-1640 medium supplemented with 10% inactivated fetal calf serum and 50 µg/mL gentamycin. The cells were maintained at 37 o C in a humidified atmosphere with 5% CO 2 and were subcultured two to three times a week.

In Vitro Analysis of P-gp Content
The content of P-gp in the MCF-7/ADR cell lysates after incubation with varying conc. (12.5-100 µM) of tested compounds 4a, 4c, 4d, 4e, 4g and 4i following exposure for 48 hr was determined using a commercial human P-gp (Permeability Glycoprotein) ELISA Kit (MBS2506188, MyBioSource Inc., San Diego, CA, USA). Absorption was recorded at 450 nm with a Spectramax Gemini fluorescence microplate reader (Molecular Devices, Sunnyvale, CA, USA) [70].

Rhodamine 123 Accumulation Assay
P-gp activity was determined by measuring intracellular accumulation of Rhodamine 123 in MCF-7/ADR cells in the absence or presence of compounds 4a, 4c, 4d, 4e, 4g and 4i, according to the commercial Rhodamine Competitive ELISA Kit (AKR-5142, Cell Biolabs Inc., San Diego, CA, USA) which provides a convenient method for the detection of total rhodamine in extracts from cells [71]. Briefly, resistant cancer cells were harvested, washed twice and counted. After dilution to 1 × 10 6 cells/mL in each well of 6 well plates, 1 mL of fresh media containing different concentrations of compounds 4a, 4c, 4d, 4e, 4g and 4i were added and incubated at 37 • C for 4 h in an atmosphere containing 5% CO 2 . Subsequently, 5.25 µM of Rho123 was added to each well and the wells were incubated for another 30 min at 37 • C. Finally, cells were washed and lysed as described before, and intracellular quantification levels of Rhodamine 123 were further analyzed according to the ELISA protocol kit. Absorbance at 450 nm of each well was measured using a Spectramax Gemini fluorescence microplate reader (Molecular Devices, Sunnyvale, CA, USA). The total content of Rhodamine in each sample was determined by comparison with a Rhodamine standard curve.

Molecular Docking
We have used a program similar to that used by Jaguar et al. [72] to generate all possible tautomeric and stereo-isomeric stats for the structures. Crystal structures of Pglycoprotein were taken from the protein data bank and bonded with 5-florouracil as the reference drug. All ligands were imported into Ligprep module and redocked into the appropriate binding sites using Glide's module. The Glide tool was applied to perform the molecular docking, then a grid for protein was charged using the default aspects of force field. The (SP) scoring function was produced to study the binding affinity, and then charged with a Charm force field. The low root square devotion RMSD score was utilized to obtain the other poses. To draw, the Schrodinger builder was applied.
The introduction of halogen atoms in 1-position on the pendant phenyl group plays a more important role in enhancing antitumor activities than methoxy groups, according to structure activity relationships (SARs).