Introduction

The aim of the study was to evaluate the anticancer activities of ethyl 3,5-diphenyl-2-cyclohexenone-6-carboxylate derivatives in HCT116 human colon cancer cells. Chalcone contains two phenyl rings connected by an α,β-unsaturated carbonyl group. More than a thousand chalcones were reported in PubChem (https://pubchem.ncbi.nlm.nih.gov/) and several tens of thousands could be searched in SciFinder (https://scifinder.cas.org/). The increased research interest in chalcones stems from its diverse biological activities, including anti-inflammatory, antioxidant, antifungal, antimicrobial, and antimalarial, among others [1,2,3,4,5,6]. Mdl 27,048 (2′,5′-dimethoxy-4-(dimethylamino)chalcone) inhibited tubulin polymerization [7]. Butein (3,4,2′,4′-tetrahydroxychalcone) inhibited inhibitor of nuclear factor kappa-B kinase subunit beta (IKKβ), which resulted in the translocation of the nuclear factor kappa-light-chain enhancer of activated B cells [7]. Xanthohumol (4,2′,4′-trihydroxy-6′-methoxy-3′-(methylbut-2-enyl)chalcone) inhibited aromatase, the inhibitors of which were used for the treatment of breast cancer [8]. Xanthoangelol (2′,4′,4-trihydroxy-3′-((2E)-3,7-dimethylocta-2,6-dienyl)chalcone) inhibited aurora A and B kinases, which were known oncoproteins [9]. In our previous studies, 2-hydroxy-4-methoxy-2′,3′-benzochalcone was found to inhibit tubulin polymerization [10] and 2′-hydroxy-2,4,6-trimethoxy-5′,6′-naphthochalcone induced G2/M cell cycle arrest and apoptosis [11]. An anticancer effect was common among these chalcones. The α,β-unsaturated carbonyl group in chalcones acted as a Michael acceptor, which can form a covalent bond with the thiol group in cysteine or glutathione. The formation of the covalent bond limited the elevation of the glutathione level. In our previous research, we demonstrated that 5′,3,5-trimethoxychalcone killed cancer cells selectively via the generation of reactive oxygen species (ROS) [12]. In cancer cells, increased levels of ROS led to cell death; however, in normal cells, these species stimulated cellular proliferation [13, 14]. Besides, the Michael acceptor in chalcones activated Kelch-like ECH-associated protein 1 [15]. Therefore, the authors of that study attempted to modify the chalcone, retaining the Michael acceptor. A study reported that 3,5-diaryl-2-cyclohexenone attenuated heart failure in a zebra fish model [16]. Figure 1 (left and right) presents the structures of 5′,3,5-trimethoxychalcone and 3-phenyl-1-oxocyclohex-2-ene (3-phenyl-2-cyclohexenone), respectively. The boxes in the structures indicate the Michael acceptor calyxol (ethyl 3-methyl-5-pentyl-2-cyclohexenone-4-carboxylate), which has been reported to have bed bug repellency [17]. As mentioned above, because compounds containing cyclohexenone or/and carboxylate moiety showed various biological activities, we designed derivatives of ethyl 3,5-diphenyl-2-cyclohexenone-6-carboxylate (Fig. 2).

Fig. 1
figure 1

Structures of (left) 5′,3,5-trimethoxychalcone and (right) 3-phenyl-1-oxocyclohex-2-ene (3-phenyl-2-cyclohexenone); the boxes denote the Michael acceptor

Full size image
Fig. 2
figure 2

Structure of ethyl 3,5-diphenyl-2-cyclohexenone-6-carboxylate

Full size image

Of the multiple methods that can be used to screen anticancer compounds, we opted for a clonogenic long-term survival assay, given its ability to distinguish small differences in the antitumor activity caused by compounds with similar structures [18]. To verify the biological activity of the cyclohexenone derivatives, we determined the activation of the apoptosis regulatory protein caspases [19], through immunoblotting and immunofluorescence microscopy in HCT116 colon cancer cells. Compounds containing a cyclohexenone moiety have been reported to inhibit acetylcholinesterase (AChE) [20]. In this study, we predicted that ethyl 3,5-diphenyl-2-cyclohexenone-6-carboxylate derivative binds to AChE using an in silico docking experiments. The druggability of the cyclohexenone derivatives was evaluated based on ligand efficiency. Our results suggest that ethyl 3,5-diphenyl-2-cyclohexenone-6-carboxylate derivatives inhibit AChE activity and trigger apoptosis in HCT116 colon cancer cells.

Materials and methods

Twenty-one derivatives of ethyl 3,5-diphenyl-2-cyclohexenone-6-carboxylate (listed in Table 1) were prepared according to the previous report [21]. A clonogenic long-term survival assay was performed in HCT116 human colon cancer cells, as previously described [22]. HCT116 cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA) and maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% (v/v) heat-inactivated fetal bovine serum (HyClone, Logan, UT, USA). The cells were seeded into 24-well tissue culture plates (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA) at 3 × 103 cells per well. After attachment, the cells were exposed to different concentrations of the derivatives (0, 5, 10, 20, and 40 μM) for 7 days, followed by fixation in 6% (w/v) glutaraldehyde and staining with 0.1% (w/v) crystal violet [23]. For all of these analyses, previously reported methods were followed [25].

Table 1 List of cyclohexenone derivatives
Full size table

Immunoblotting was performed as described previously with minor modifications [24]. Briefly, HCT116 cells were lysed in a buffer containing 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES; pH 7.2), 1% Triton X-100, 10% glycerol, 150 mM sodium chloride (NaCl), 10 μg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride (PMSF). Whole-cell lysates were electrophoresed on a 10% sodium dodecyl sulfate (SDS)-polyacrylamide gel and transferred onto nitrocellulose membranes (Bio-Ras, Richmond, CA, USA). After blocking the membranes with 10% skim milk in Tris saline buffer (10 mM Tris–HCl, pH 8.0, 150 mM NaCl, and 0.1% Tween 20) for 1 h at 25 ℃, primary antibodies were added and the membranes were incubated overnight at 4 ℃. Primary antibodies against cleaved caspase-9, cleaved caspase-7 (Asp198), cleaved caspase-3 (Asp198), and poly(ADP-ribose) polymerase (PARP) were obtained from Cell Signaling Technology (Beverly, MA, USA), and antibodies against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). After washing the membranes thrice with Tris saline buffer, horseradish peroxidase (HRP)-conjugated secondary antibodies (Cell Signaling Technology) were added and incubated for 4 h at 25 ℃. After washing the membranes five times with TBST, the blots were developed using an Amersham ECL Western Blotting Detection Kit (GE Healthcare Life Science, Chicago, IL, USA).

An in vitro acetylcholinesterase enzyme activity assay was performed using an acetylcholinesterase assay kit (fluorometric-green; ab138872; Abcam, Cambridge, UK). Thiocholine produced from the hydrolysis of acetylthiocholine by AChE can be quantified using thiol green indicator. The intensity of thiol green indicator was measured on an FS-2 fluorescence spectrophotometer (Scinco, Seoul, Korea) fitted with a xenon lamp and bandwidth-adjustable filters for the excitation and emission wavelengths. Samples of all 21 derivatives were dissolved in dimethylsulfoxide and adjusted to 1, 5, 10, 50, and 80 μM concentrations. Donepezil (used as a reference drug) solutions were prepared at 0.001, 0.01, 0.1, 1, 5, and 10 μM concentrations. The excitation and emission wavelengths were 490 and 520 nm, respectively. IC50 values were determined using SigmaPlot, following the manufacturer’s protocol.

Immunofluorescence staining was conducted as described previously [26]. HCT116 cells were cultured on coverslips and treated with derivative 21 for 12 h. After fixing the cells with 4% paraformaldehyde and permeabilizing them with 0.1% (v/v) Triton X-100, the cells were incubated with primary antibodies against α/β-tubulin and cleaved caspase-7 for 2 h at 25 ℃. It was probed with Alexa Fluor 488- (for α/β-tubulin; green signal) and Alexa Fluor 555-conjugated (for cleaved caspase-7; red signal) secondary antibodies for 30 min at 25 ℃. Nuclear DNA was counterstained with 1 μg/mL Hoechst 33,258 for 10 min (blue signal). Fluorescence was examined under an EVOS FL fluorescence microscope (Advanced Microscopy Group; Bothell, WA, USA).

As a three-dimensional (3D) structure of AChE, 4ey7.pdb deposited in the protein data bank (PDB) was selected [27] for the in silico docking experiments conducted on a Linux PC using the Sybyl program (Tripos, St. Louis, MO, USA). The experimental procedures followed have been described previously in detail [28].

Results and discussion

The 21 cyclohexenone derivatives used in this study contained 3 substituents, including 3,5-diphenyl rings and ethyl 6-carboxylate (Fig. 2). The 3,5-diphenyl group was substituted with 1-naphthalene or 2-naphthalene. The clonogenic assay shows that most of the derivatives inhibited the clonogenic ability of HCT116 cells at concentrations of 20–40 μM (Fig. 3). The 1-naphthalenyl substituents exhibited better inhibitory activity on the clonogenicity of HCT116 cells than the 2-naphthalenyl substituents, and the number and position of the substituents attached to the 3-phenyl ring, including the hydroxy and methoxy groups, did not affect the clonogenicity of HCT116 cells (Table 1). These data suggest that cyclohexenone derivatives used in this study exhibit anticancer properties (Fig. 4).

Fig. 3
figure 3

Effect of 21 cyclohexenone derivatives on the clonogenicity of HCT116 cells

Full size image
Fig. 4
figure 4

Effect of 21 cyclohexenone derivatives on AChE activity in vitro. Bars represent mean ± standard deviation (n = 3). IC50, half-maximal enzyme inhibitory concentration; Donepezil, an AChE inhibitor reference compound.

Full size image

AChE is a serine hydrolase found at cholinergic synapses and neuromuscular junctions. It plays a crucial role in terminating neuronal transmission by hydrolyzing acetylcholine into acetic acid and choline [30]. In addition to the classical function of acetylcholine hydrolysis, AChE also exerts non-cholinergic functions such as the promotion of apoptosis in HeLa cells [31] and fibroblasts [32]. However, several studies reported that acetylcholine induces the proliferation of gastric cancer cells [33] and stimulates lung cancer growth [34, 35]. The enzymatic activity of AChE is enhanced in meningiomas, gliomas, and vestibular nerve schwannomas [36,37,38,39]. Thus, AChE functions differently in different tissue types. In human colon cancer cells, the stimulation of the muscarinic cholinergic M3 receptor induced cellular proliferation [40] and stimulated migration and invasion [41, 42]. In addition, acetylcholine is synthesized and mediates the autocrine stimulation of colon cancer cell proliferation [43]. These findings suggest that AChE inhibitors may be useful chemotherapeutic agents against colon cancer.

As compounds containing a cyclohexenone moiety can inhibit AChE [44], we tested whether cyclohexenone derivatives inhibit AChE enzyme activity. In vitro AChE enzyme activity assay shows that the half-maximal enzyme inhibitory concentration (IC50) values ranged from 0.93 to 133.12 μM (Fig. 5 and Table 1). Among them, derivative 21 exhibited the best AChE inhibitory activity (IC50 = 0.93 μM). The IC50 value of donepezil, 2-((1-benzylpiperidin-4-yl)methyl)-5,6-dimethoxy-2,3-dihydro-1h-inden-1-one, known as an AChE inhibitor, was determined to be 0.13 μM under the same experimental conditions. The trade name of donepezil is Aricept, which is used in the treatment of Alzheimer’s disease [45].

Fig. 5
figure 5

Effect of derivative 21 on the activation of the caspase cascade. a HCT116 cells were treated with 50 μM derivative 21 for various durations of time (0–24 h), and whole-cell lysates were subjected to immunoblotting using antibodies specific for the cleaved caspases. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control to ensure the same amount of protein loading. b HCT116 cells cultured on coverslips were treated with 50 μM derivative 21 for 12 h. The cells were fixed and incubated with antibodies against α/β-tubulin and cleaved caspase-7 for 2 h, followed by incubation with AlexaFluor 488- (green signal) and AlexaFluor 555-conjugated (red signal) secondary antibodies for 30 min. Nuclear DNA was stained with 0.1 μg/mL Hoechst 33,258 for 10 min (blue signal). Arrows indicate the apoptotic nuclear fragments. Fluorescence-positive cells were examined under an EVOS FL® fluorescence microscope. Dotted boxes are enlarged on the right. Size bars, 50 μm

Full size image

Caspases are cysteine-dependent proteases that mediate the apoptosis triggered by various anticancer agents [19]. Caspases are activated via proteolytic cleavage during the initiation (caspase-9) and execution stages (caspase-3 and -7) of apoptosis [19]. To investigate whether the inhibition of AChE induces apoptosis, we treated HCT116 cells with 50 μM derivative 21, which showed the highest AChE inhibitory activity, and examined the activation of caspases through immunoblotting. Upon treatment with derivative 21, the proteolytic fragments of the initiator caspase (caspase-9) and executioner caspases (caspase-3 and -7) increased in a time-dependent manner (Fig. 5a). Poly(ADP-ribose) polymerase (PARP) is a substrate protein of caspase-3 and -7 [29]. The extent of the proteolytic cleavage of PARP was also increased as a function of time, following treatment with derivative 21 (Fig. 5a). Immunofluorescence microscopy also showed high levels of cleaved caspase-7 upon treatment with derivative 20 (Fig. 5b). Notably, fragmented nuclei, a characteristic of apoptotic cells, appeared after treatment with derivative 21. These data suggest that derivative 21 triggers apoptosis via the activation of the caspase cascade in HCT116 colon cancer cells.

To elucidate the binding mode of AChE and cyclohexenone derivatives, we conducted in silico docking experiments using derivative 21. Multiple 3D structures of AChE are available in the PDB. Even the PDB structures 1f8u.pdb and 4bdt.pdb contained the highest number of residues, with resolutions of 2.90 Å and 3.10 Å, respectively [46, 47]; they did not contain donepezil as the ligand, which was used as a reference compound in the current in vitro AChE enzyme assay. Therefore, 4ey7.pdb containing donepezil as its ligand and with a resolution of 2.35 Å was selected for in silico docking [27]. Although human AChE consists of 614 amino acids, 4ey7.pdb contains 542 amino acid residues (Gly33–Thr574), with homodimer polypeptides A and B. Since chain A contained more unmodeled residues than chain B, the latter was chosen for docking purposes. The apo-protein of chain B, prepared using the Sybyl program, was subjected to energy minimization. The root mean square deviation between the crystallographic structure and energy minimized apo-protein was 0.22 Å. Residues in the binding site were determined via LigPlot analysis [48]: Try103, Trp117, Gly152, Tyr155, Glu233, Ser234, Trp317, Ser324, Phe326, Tyr368, Phe369, Tyr372, and His478. The 3D structure of derivative 21 was determined based on the 3D structure of 5-pentafluorophenyl-3-phenyl-2-cyclohexenone available at PubChem. The Sybyl program provides a flexible docking method. To confirm the accuracy of the docking procedures, the donepezil ligand of 4ey7.pdb that was docked onto the apo-protein of AChE chain B was observed to dock well (Additional file 1: Fig. S1). The Sybyl program generated 30 ligand–protein complexes owing to the 30 repeated iterations. The binding energy obtained from the docking results ranged between 13.55 and −  10.62 kcal/mol. Likewise, in silico docking for derivative 21 revealed the binding energy to range from − 11.96 to − 10.91 kcal/mol. The derivative 21 ligands generated by 30 iteraton were docked into the apo-AChE protein well (Additional file 1: Fig. S2). The complex with the lowest binding energy showed the best docking pose (Additional file 1: Fig. S3) and was thus subjected to analysis using LigPlot (Additional file 1: Fig. S4). Nine residues, including Trp286, Leu289, Glu292, Val294, Arg296, Phe297, Phe338, Tyr341, and Gly342, showed hydrophobic interactions, and two residues, Ser293 and Phe295, formed hydrogen bonds (H-bonds). H-bonds were observed between the amino proton of the peptide bond of Ser293 and the ketone oxygen of carboxylic acid (3.11 Å) and between the amino proton of the peptide bond of Phe295 and the oxygen of the methoxy group attached to the phenyl ring (2.89 Å). The 3D image of the binding site of the derivative 21–AChE complex was generated using PyMol program (The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC. Portland, OR, USA) (Fig. 6). The reason for the low binding energy of the derivative 21–AChE complex compared to the donepezil–AChE complex may be related to their IC50 values, i.e., 0.93 μM and 0.13 μM, respectively. Besides, while donepezil was docked inside the binding site (Additional file 1: Fig. S1), derivative 21 was docked at the entrance of the binding site (Additional file 1: Fig. S3).

Fig. 6
figure 6

Three-dimensional (3D) image of the binding site of the derivative 21-acetylcholinesterase complex generated by PyMol (The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC. Portland, OR, USA)

Full size image

The molecular weights of the 21 cyclohexenone derivatives ranged between 386 and 476 Da (Table 1). While 18 derivative compounds were novel, derivatives 1, 10, and 20, though not published, were registered with the Chemical Abstract Service (https://www.cas.org/), vide registration numbers, 1632161-14-8, 1194722-11-6, and 52220-43-6, respectively. Lead optimization, necessary to obtain novel and active compounds, resulted in increased molecular weights, with a subsequent enhancement in hydrophobicity as well [49]. Hydrophobicity can be predicted by the logP values. In this study, the logP values that were calculated using the Sybyl/MOLCAD module, ranged from 4.10 to 4.48 (Table 1). Because highly lipophilic compounds have low solubility, in vivo experiments, including the preclinical phase, are challenging to perform. Therefore, this requires an additional delivery system.

Ligand efficiency (LE) is the ratio of the ligand affinity for the target molecules to the number of heavy atoms except for hydrogen atoms [49], which has been used to rank hit compounds for drug development. LE can be obtained from the equation as follows:

$${\text{LE }}\; = \; - \;\Delta {\text{G}}/{\text{HA}}$$

where ΔG and HA denote the Gibbs free energy change associated with the binding of the ligand–protein complex and the number of non-hydrogen atoms, respectively [50]. The LE value for the derivative 21–apo-AChE complex obtained using this equation was 0.36 (ΔG = − 11.88 kcal/mol, HA = 33), which falls into the criteria of LE values for drug-like compounds [50]. Therefore, the cyclohexenone derivatives obtained in this research can be considered potent chemotherapeutic agents. However, the study has some limitations. The limited number of tested derivatives in this study may not provide sufficient evidence for the chemotherapeutic potential. Further in vivo studies are needed to confirm the correlation between the cyclohexenone derivative-induced AChE inhibitory activity and anticancer property.