Natural product-inspired synthesis of coumarin–chalcone hybrids as potential anti-breast cancer agents

Twelve novel neo-tanshinlactone–chalcone hybrid molecules were constructed through a versatile methodology involving the Horner–Wadsworth–Emmons (HWE) olefination of 4-formyl-2H-benzo [h]chromen-2-ones and phosphonic acid diethyl esters, as the key step, and evaluated for anticancer activity against a series of four breast cancers and their related cell lines, viz. MCF-7 (ER + ve), MDA-MB-231 (ER-ve), HeLa (cervical cancer), and Ishikawa (endometrial cancer). The title compounds showed excellent to moderate in vitro anti-cancer activity in a range of 6.8–19.2 µM (IC50). Compounds 30 (IC50 = 6.8 µM and MCF-7; IC50 = 8.5 µM and MDA-MB-231) and 31 (IC50 = 14.4 µM and MCF-7; IC50 = 15.7 µM and MDA-MB-231) exhibited the best activity with compound 30 showing more potent activity than the standard drug tamoxifen. Compound 30 demonstrated a strong binding affinity with tumor necrosis factor α (TNF-α) in molecular docking studies. This is significant because TNFα is linked to MCF-7 cancer cell lines, and it enhances luminal breast cancer cell proliferation by upregulating aromatase. Additionally, virtual ADMET studies confirmed that hybrid compounds 30 and 31 met Lipinski’s rule; displayed high bioavailability, excellent oral absorption, favorable albumin interactions, and strong penetration capabilities; and improved blood–brain barrier crossing. Based on the aforementioned results, compound 30 has been identified as a potential anti-breast cancer lead molecule.


Introduction
Breast carcinoma is the most commonly diagnosed cancer and the leading cause of death from cancer disease globally.In the year 2020, approximately 23 million new cases of breast cancer (11.7% of all cancer cases) were estimated worldwide (Siegel et al., 2020;Sung et al., 2021;Saquib et al., 2023).Despite the achievements and advances made in the chemotherapy of cancers during the past 20 years, the emergence of drug resistance continues to be a major issue in cancer treatment.Increasing prevalence of drug resistance combined with issues of tissue selectivity and toxicity in cancer chemotherapy has necessitated further research toward the development of less toxic and more effective therapeutic antibreast cancer agents (Saquib et al., 2019).The search for smallmolecule inhibitors (SMIs) based on novel molecular scaffolds that exhibit unique medicinal properties is largely inspired by biologically relevant molecules, primarily natural products (NPs).The structural diversity of NPs provides a rich source of molecular scaffolds that can be used as starting points for the design and discovery of new SMIs.By studying the structural features of NPs, researchers can identify new therapeutic agents with improved pharmacokinetic properties, increased potency, and reduced toxicity.However, the full potential of NPs in the design of new scaffolds for drug discovery is currently limited by the scarcity of guiding synthetic strategies available for this purpose.In this milieu, molecular hybridization can be a very useful tool (Li and Vederas, 2009;Wetzel et al., 2011).This approach involves combining two or more pharmacophoric units to obtain novel compounds with enhanced therapeutic properties and is finding increasing applications in the fields of drug discovery and medicinal chemistry (Meunier, 2008;Bosquesi et al., 2011;Saquib et al., 2013;Tukulula et al., 2013;Švenda et al., 2015) (Figure 1) (Tietze et al., 2003).
Several naturally occurring and synthetic coumarin derivatives have been reported to exhibit potent anti-cancer activity including anti-breast cancer activity (Musa et al., 2008;Küpeli Akkol et al., 2020).Neo-tanshinlactone, a benzocoumarin isolated from the underground stem of Salvia miltiorrhiza, inhibited the proliferation of MCF-7 cell lines more effectively than the antibreast cancer drug tamoxifen (Wang et al., 2004;Lin et al., 2016).Likewise, chalcone is an important pharmacophore widely distributed in many natural products.Various natural and synthetic chalcones have been reported to show excellent biological activities including anticancer activity (Malek et al., 2011;Zhang et al., 2013;Zhuang et al., 2017).Moreover, a naturally occurring coumarin-chalcone hybrid isolated from the leaves of Cyclosorus parasiticus have been found to be cytotoxic (Quadri-Spinelli et al., 2000;Wei et al., 2013).
The interesting pharmacological properties of the neotanshinlactone and the chalcone scaffolds inspired us to design the neo-tanshinlactone-chalcone hybrid molecules (coumarin-chalcone hybrids) by the de novo combination of both these scaffolds onto one template (Figure 3).

In vitro antiproliferative activities of the synthesized compounds
We conducted an in vitro MTT assay to evaluate the anti-cancer activity of the synthesized neo-tanshinlactone-chalcone hybrid compounds, 26-37, against four human cancer cell lines, namely, MCF-7 (breast cancer), MDA-MB-231 (breast cancer), Ishikawa (endometrial cancer), and HeLa (cervical cancer).The toxicity of the hybrid molecules on the normal human embryonic kidney cell line HEK-293 was also evaluated.
Similarly, compound 31 also exhibited potent inhibitions against MCF-7 and MDA-MB-231 cell lines with IC 50 values of 14.4 µM and 15.7 µM, respectively, while compound 26 exhibited anti-cancer activity with IC 50 values of 17.2 µM and 16.6 µM, respectively, in both of these cell lines.Molecules 34, 35, 36, and 37 having chloro substituents on the aroyl ring were found inactive against MCF-7 and MDA-MB-231 cell lines (IC 50 > 20 µM); however, compounds 36 and 37 showed good antiproliferative activity against Ishikawa and HeLa cell lines.Molecule 36 showed anti-cancer activity against Ishikawa and HeLa cell lines and with IC 50 values of 19.2 µM and 16.8 µM, respectively, while compound 37 exhibited inhibition against Ishikawa and HeLa cell lines with IC 50 values of 18.6 µM and 17.2 µM, respectively.Molecules 29 and 32 having unsubstituted aroyl rings were found inactive (IC 50 > 20 µM) against all the four cancer cell lines.The results of the assay showed that the synthesized compounds had significant anti-cancer activity against all the four cell lines.

Analysis of molecular dynamic simulation
After the analysis of the in vitro and docking studies, we found that compound 30, as the most active derivative of neo-tanshinlactone, inhibited the cell proliferation of cancer cells via the suppression of the TNF-α, MAPK, and Nf-kB signaling pathways.Because of the highest binding, we performed MD simulation on compound 30 to study the stability within binding pockets of TNF-α target protein.A 100-ns trajectory was used to extract out RMSD, RMSF, RyG, and the number of hydrogen bonds of the ligand-protein complex by applying the OPLS3 force field in the molecular system.In this experiment, we found that the RMSD value of the ligand varies from 0.512 Å to 2.010 Å (Table 3A) and it exhibited the conformational rotation in the active site of the protein.The RMSD value of the protein varies from 1.162 Å to 2.402 Å (Table 3).Initially, it showed an RMSD value 2.402 for 20 ns time and then attained the equilibrium throughout the simulation.The RMSF value of Cα protein deviated from 0.363 Å to 3.987 Å (Table 3A), but the residue Asn39 from the chain E fluctuated to 3.987 Å, which is attributed to the stability for the whole simulation (Khan et al., 2022).In order to optimize the ligand properties, 30 was found to deviate up to 18.30 ns, which had shown RMSD values in the range of 0.62 Å to 1.38 Å and then attained equilibrium at 0.41 Å.The rGyr value of the ligand was calculated from 4.295 Å to 4.850 Å, followed by the achievement of the equilibrium point at 4.5 Å.Similarly, MolSA, SASA, and PSA were computed for the ranges 369.444Å 2 -393.097Å 2 , 0.000 Å 2 -11.171Å 2 , and 118.929Å 2 -146.046Å 2 .As shown in Figure 11, these properties of compound 30 showed that it remains stable throughout the MD simulation.The protein-ligand interactions of the stable 30-TNF-α complex were analyzed by a histogram, as shown in Figure 11F.The protein-ligand interactions were distinguished by including hydrogen, ionic, and hydrophobic bonds, and water bridges.We observed that compound 30 showed five hydrogen bonds with chain D: Tyr151, chain E: Gln61, chain E: Gly122, chain E: Tyr151, and chain F: Gly122 amino acids of the TNF-α target (Table 4).On the other hand, throughout the MD simulation, it formed a very stable intermolecular hydrogen bond with chain E: Tyr151, which contributed to more than 99% stability of the compound 30-TNF-α complex.Despite the hydrogen bonds, compound 30 also showed 14 hydrophobic interactions with chain D: Leu57, chain D: Tyr59, chain D: Tyr119, chain D: Tyr151, chain D: Ile155, chain E: Leu57, chain E: Tyr59, chain E: Tyr119, chain E: Val123, chain F: Leu57, chain F: Tyr59, chain F: Tyr119, chain F: Val123, and chain F: Ile155 residues (Table 3B) (Abdalla et al., 2021;Kumar et al., 2022).

Pharmacological parameters of compounds
Preliminary screening of the hybrid compounds showed that none of the compounds violated the rule of five (Lipinski's rule), and all of them showed high bioavailability.Consequently, we studied the pharmacological properties of compounds 30 and 31, which showed the highest molecular docking score, in detail.Both the compounds exhibited 100% human oral absorption and good interactions with serum albumin 0.24 and 0.26 within the range (−1.5 to +1.5), respectively.Furthermore, they demonstrated high penetration capabilities with values of 786.69 nm/s and 786.00 nm/s against Caco, and 381.70 nm/s and 381.34 nm/s against MDCK cells, respectively.Moreover, compounds 30 and 31 showed a significantly improved capability to cross the blood-brain barrier with values of −1.11 and −1.09, which were lower than the specified criterion from −3.0 to 1.2 (Table 4).Considering the aforementioned results, we can say that compounds 30 and 31 have all the attributes to be potent therapeutic agents (Kumar et al., 2023).
Frontiers in Pharmacology frontiersin.orgthe Claisen-Schmidt reaction was used instead of the present strategy.However, our efforts to couple the β-aryl-βketophosphonates with the 4-formyl-2H-benzo [h]chromen-2ones to obtain the target hybrid molecules using the Claisen-Schmidt reaction were not successful.Considering the potential usefulness of the coumarin-chalcone hybrids as novel anti-breast cancer compounds, we are now concentrating our efforts toward finding a new synthetic protocol which will enable us to synthesize newer derivatives of the aforementioned hybrid molecules and evaluate their anti-breast cancer activity in our search for more potent and less toxic anti-breast cancer agents.

Conclusion
In summary, a library of 12 coumarin-chalcone hybrid compounds, 26-37, were constructed, based on the molecular hybridization concept, via a practical and effective synthetic route which involved the HWE coupling of 4-formyl-2H-benzo [h] chromen-2-ones with phosphonic acid diethyl esters.The in vitro anti-cancer activity of the synthesized hybrid molecules was evaluated using an MTT assay, wherein they were found to exhibit potent to moderate anti-cancer activity against a panel of four cancer cell lines, including two breast cancer (MCF-7, ER + ve and MDA-MB-231, ER-ve), one cervical cancer (HeLa), and one endometrial cancer (Ishikawa) cell lines.Among these, molecules 30 and 31 were found to be the most potent, with compound 30 demonstrating superior activity against both ER + ve (IC 50 = 6.8 µM, MCF-7) and ER-ve (IC 50 = 8.5 µM, MDA-MB-231) breast cancer cell lines as compared to the standard drug tamoxifen.The toxicity of the molecules, 26-37, was also evaluated on the normal human embryonic kidney cell line HEK-293, and interestingly, none of the screened molecules exhibited any toxicity against normal HEK-293 cells.Compound 30 also demonstrated a strong binding affinity with the tumor necrosis factor α (TNF-α) in molecular docking studies.This is significant because TNFα is associated with MCF-7 cancer cell lines.Virtual ADMET studies validated the compliance of hybrid compounds 30 and 31 with Lipinski's rule.Moreover, they exhibited high bioavailability, excellent oral absorption, favorable albumin interactions, strong penetration capabilities, and improved blood-brain barrier crossing, indicating their potential as potent therapeutic agents.Thus, compound 30 could be considered a potential anti-breast cancer lead molecule, possibly by targeting TNFα.

Chemistry
All the chemicals were procured from Across Organics and Sigma-Aldrich, and were used without further purification.The IR spectra of the compounds were recorded using a Perkin-Elmer Spectrum GX FTIR spectrometer. 1 H NMR and 13 C NMR spectra were recorded using the Bruker DRX-300 (300 MHz for 1 H and at 75 MHz for 13 C) or DPX-200 (at 50 MHz for 13 C) spectrometer using CDCl 3 , DMSO-d 6 , or TFAd 1 (see the Supporting Information).Chemical shifts (δ) are reported in parts per million (ppm), using TMS as an internal standard.ESI-MS spectra were recorded using an LCQ advantage ion-trap mass spectrometer (Finnigan Thermo Fisher Scientific), and highresolution mass spectra (ESI-HRMS) were recorded using an Agilent 6520 ESI-QTOP mass spectrometer.9.1.1General procedure for the synthesis of 2bromo-1-(aryl) ethan-1-ones (7)(8)(9)(10)(11)(12) A solution of bromine (7.50 mmol; 1.0 equiv), dissolved in Et 2 O (15 mL), was added to the stirred solution of the respective substituted acetophenones (7.5 mmol) in Et 2 O (15 mL) in a 50-mL round bottom flask at 0 °C, and the mixture was stirred for 2 h at room temperature.After the completion of the reaction, it was quenched with the saturated NaHCO 3 solution (100 mL) and extracted with Et 2 O (3 × 40 ml).The combined organic layers were dried over anhydrous Na 2 SO 4 and concentrated in vacuo to obtain the crude mixture which was subsequently crystallized from EtOH to obtain the pure compounds 9 and 10.Compounds 7,8,11,and 12, on the other hand, are commercially available.

Cell proliferation assay (MTT assay) in cancer cell lines and normal cell lines
The synthesized hybrid compounds were tested for their anticancer activity using the MTT reduction assay on 2.5 × 10 3 cells per well that were seeded in 100 µL DMEM supplemented with 10% FBS in each well of 96-well microculture plates.After seeding, the cells were incubated for 24 h at 37 °C in a CO 2 incubator.The required concentrations of the compounds were achieved by diluting the solutions in the culture medium.The media was removed from the wells after 48 h, and 100 µL of MTT (0.5 mg/mL) was added to each well.The plates were then incubated for an additional 3 h.After carefully removing the supernatant from each well, the formazan crystals were dissolved in 100 µL of DMSO.The absorbance was then recorded at a wavelength of 540 nm (Saquib et al., 2021;Gupta et al., 2022;Iqbal et al., 2023)

Molecular docking studies
Molecular docking is used to predict the possible binding abilities of ligands within the pockets of amino acid residues of target protein receptors.To find out the possible binding modes, the calculation of docking scores was carried out for all synthesized compounds, 26-37, against four cell proliferating proteins, namely, NF-kβ, TNF-α, MAPK, and JNK, using the Glide module of Maestro version 12.6.144(Schrödinger 2020-4 LCC, New York, United States) software.The chemical structures of compounds, 26-37, were drawn using ChemDraw Professional 15.0 software, and the 2D structures were saved in the mol format (Table 1).Ligands were prepared using the LigPrep module of Maestro version 12.6.144(Schrödinger 2020-4 LCC, New York, United States) software.The minimization of all ligands corrected the bond length and angle, and generated the possible conformers by using the OPLS force field.All the feasible protonation and ionization states were itemized for each compound using Epik pH 7.0±2.0and then saved in the Maestro format (Fadaka et al., 2022).The crystal structure of protein receptors, namely, TNF-α, (PDB ID 6X81), JNK (1JNK), MAPK (1A9U), and NF-kB (1SVC), having the resolution of 2.81 Å, were obtained from the RCSB PDB database (https://www.rcsb.org/)and then prepared by using the protein preparation wizard module of Maestro version 12.6.144(Schrödinger 2020-4 LCC, New York, United States) software.The refinement of the protein structure was carried out by the removal of the heteroatom and water molecules, followed by the assignment of the bond order with added hydrogen atoms using the OPLS force field and minimized the receptor (Pattar et al., 2020).In order to identify the top-ranked potential receptorbinding sites in protein, the active sites were predicted for ligand binding to the receptor using the sitemap module of Maestro version 12.6.144(Schrödinger 2020-4 LCC, New York, United States) software (Halgren, 2009).The grid box was centroid over the active site of the protein, which was predicted by the sitemap with the van der Waals scaling factor of 1. 0 Å, and the partial charge cut-off was 0.25 Å (Sharma et al., 2016).For molecular docking, the softened potential was considered by the van der Waal radii scaling factor of 0.80 Å with a 0.15-Å partial charge cut-off for non-polar parts of ligands.After docking, we saved the best pose of the ligand-protein complex based on the docking scores (Ansari et al., 2022).

Molecular dynamic simulation
The molecular dynamic simulation study was conducted to check the stability of the ligand-protein complex.The MD simulation was implemented using the Desmond module of Maestro version 12.6.144(Schrödinger 2020-4 LCC, New York, United States) software.The best pose of the docked ligand-protein complex was immersed into the TIP3P water solvent model and generated the orthorhombic boundary with the shape size of 10 Å × 10 Å X10 Å.The system was neutralized by Na + /Cl − ions at 0.15 M salt concentration, and the whole system was minimized by applying the OPLS3 force field.The MD simulation was performed in an NPT ensemble at 300 K and 1.013 bar pressure over the 100-ns simulation.To maintain the temperature and pressure, we used the Nose-Hoover chain thermostat and Martyna-Tobias-Klein barostat method, respectively.The RESPA (reversible reference system propagator algorithm) integrator was used with time to accelerate the simulation of each 2.0 fs step with the PME (particle-mesh Ewald) method for the calculation of the long-range electrostatic interaction.Finally, the energy was evaluated at every 100.0 ps interval after analyzing the simulation trajectory (Ansari et al., 2022).

Pharmacological properties
The pharmacological properties of 26-37 compounds were assessed using the QikProp v6.8 tool of Maestro 12.6.144(Schrödinger 2020-4 LCC, New York, United States) software.The QikProp v6.8 tool analyzes a number of significant factors, such as molecular weight (MW), hydrogen bond donor (HBD), hydrogen bond acceptor (HBA), polar surface area (PSA), lipophilicity (LogP), aqueous solubility (LogS), Caco cell permeability, MSCK permeability, human serum albumin (HSA), and human oral absorption (HOA), that are useful in predicting drug-like candidate properties (Mohan et al., 2022) performed the analysis, and wrote the paper.All authors contributed to the article and approved the submitted version.

FIGURE 2
FIGURE 2Design of biologically relevant natural product hybrids as anticancer agents.

FIGURE 4
FIGURE 4Outline for the synthesis of neo-tanshinlactone-chalcone hybrids.

FIGURE 9
FIGURE 9Two-dimensional (2D) views of molecular docking: (A) molecular interactions of compound 30 within the pockets of the TNF-α target, (B) molecular interactions of compound 31 within the pockets of the JNK target, (C) molecular interactions of 30 within the pockets of the MAPK target, and (D) molecular interactions of compound 30 within the pockets of the NF-kB target.

FIGURE 10
FIGURE 10Two-dimensional (2D) views of molecular docking: (A) molecular interactions of tamoxifen within the pockets of the TNF-α target, (B) molecular interactions of tamoxifen within the pockets of the JNK target, (C) molecular interactions of tamoxifen within the pockets of the MAPK target, and (D) molecular interactions of tamoxifen within the pockets of the NF-kB target.

TABLE 1
In vitro anti-cancer activity and receptor-binding affinity of hybrid compounds, 26-37, against various cancer cell lines.

TABLE 2
Docking scores of compounds, 26-37, against the target proteins which are involved in cell proliferation.
The bold values represent the IC 50 values of compounds showing significant anti-cancer activity.

TABLE 3 A
) MD simulation parameters of the 30-TNF-α complex and B) ligand interaction with amino acids of TNF-α (PDB: 6X81.
Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS).Cells were cultured at 37 °C and 5% CO 2 .Before the experiments, cells were cultured in phenol red-free DMEM supplemented with 10% charcoal stripped fetal bovine serum.