Histone Deacetylase Inhibitory Activity and Antiproliferative Potential of New [6]-Shogaol Derivatives

Twenty newly synthesized derivatives of [6]-shogaol (4) were tested for inhibitory activity against histone deacetylases. All derivatives showed moderate to good histone deacetylase inhibition at 100 µM with a slightly lower potency than the lead compound. Most potent inhibitors among the derivatives were the pyrazole products, 5j and 5k, and the Michael adduct with pyridine 4c and benzothiazole 4d, with IC50 values of 51, 65, 61 and 60 µM, respectively. They were further evaluated for isoform selectivity via a molecular docking study. Compound 4d showed the best selectivity towards HDAC3, whereas compound 5k showed the best selectivity towards HDAC2. The potential derivatives were tested on five cancer cell lines, including human cervical cancer (HeLa), human colon cancer (HCT116), human breast adenocarcinoma cancer (MCF-7), and cholangiocarcinoma (KKU100 and KKU-M213B) cells with MTT-based assay. The most active histone deacetylase inhibitor 5j exhibited the best antiproliferative activity against HeLa, HCT116, and MCF-7, with IC50 values of 8.09, 9.65 and 11.57 µM, respectively, and a selective binding to HDAC1 based on molecular docking experiments. The results suggest that these compounds can be putative candidates for the development of anticancer drugs via inhibiting HDACs.


Results and Discussion
The natural compound [6]-shogaol (4) was isolated from ginger as previously described [27]. The immino derivatives 4a-4e were prepared by reacting [6]-shogaol (4) with hydrazine hydrate, 4-hydrazinobenzoic acid, 2-hydrazinopyridine, 2-hydrazinobenzothiazole and 2-hydrazino-2-imidazoline in ethanol at room temperature (Scheme 1). Reactions of [6]shogaol (4) with phenylhydrazines including phenylhydrazine, o-tolylphenylhydrazine, 2-methoxyphenylhydrazine, 4-methoxyphenylhydrazine, 2-fluorophenylhydrazine, 4fluorophenylhydrazine, 3-chlorophenylhydrazine, 4-chlorophenylhydrazine, 3-nitrophenylhydrazine, 4-nitrophenylhydrazine, and 4-hydrazinobenzoic acid in ethanol at 80 • C provided the pyrazole derivatives 5a-5k. All derivatives were obtained in good yields. The formations of the pyrazole rings were confirmed by the 13 C-NMR chemical shift of about 62 and 153 ppm for the C-6 and C-8 positions, respectively. Reactions of [6]-shogaol (4) with primary amines, including 2-aminothiophenol, 2-aminophenol and aniline, in ethanol at 80 • C provided the Michael-addition products 6a-6c. Interestingly, Michael immino adduct 7 was gained as the major product from reacting [6]-shogaol (4) with o-phenylenediamine in ethanol at 80 • C. The reaction showed that the amino group at the ortho-position reacts at the carbonyl group to provide a seven-membered ring. The formation of the seven-membered ring of 7 was confirmed by the chemical shift of 13  The synthesized derivatives were tested for HDAC inhibition with a commercial HDAC assay kit. The results are summarized in Table 1. All derivatives showed slightly weaker % HDAC inhibitions than the lead compound. The immino derivatives 4c and 4d showed the best HDAC inhibitory activity among the immino derivatives, with IC50 values of 61 ± 0.92 and 60 ± 0.84 µ M, respectively. The immino derivatives with aromatic groups 4a-4e showed stronger HDAC inhibitory activities than the Michael-addition products 6a-6c, 7. However, this Michael-adduct type of [6]-shogaol could become thymidylate kinase inhibitors [33]. The synthesized derivatives were tested for HDAC inhibition with a commercial HDAC assay kit. The results are summarized in Table 1. All derivatives showed slightly weaker % HDAC inhibitions than the lead compound. The immino derivatives 4c and 4d showed the best HDAC inhibitory activity among the immino derivatives, with IC 50 values of 61 ± 0.92 and 60 ± 0.84 µM, respectively. The immino derivatives with aromatic groups 4a-4e showed stronger HDAC inhibitory activities than the Michael-addition products 6a-6c, 7. However, this Michael-adduct type of [6]-shogaol could become thymidylate kinase inhibitors [33]. To study the structure-activity relationship (SAR), phenyl pyrazole derivatives with various substitution groups at the aromatic region, such as methyl, methoxy, fluoro, chloro, nitro, and carboxyl groups, were evaluated as HDAC inhibitors. The substitution phenylpyrazoles 5b-5h showed slightly weaker % HDAC inhibition than the phenylpyrazole 5a. Interestingly, the para-substitution derivatives showed highly potent HDAC inhibitors than orthoand meta-substitution derivatives. The p-nitrophenyl, pyrazole 5j, and p-carboxylphenyl pyrazole 5k derivatives were the most potent inhibitors among the pyrazole derivatives with IC 50 values of 51 ± 0.82 and 65 ± 1.12 µM, respectively.
The most potent derivatives, 4c, 4d, 5j and 5k, were further investigated as the isoformselective inhibitors of the isoforms class I HDACs (HDAC1, HDAC2, HDAC3 and HDAC8). The results from the molecular docking experiment are shown in Table 2. The immino derivative 4d showed a better binding affinity with HDAC1 and HDAC3 isoforms than TSA. The major interaction of 4d and HDAC1 ( Figure 2) consisted of two hydrogen bonds between 4d and Phe205 (2.9 Å) and Leu271 (2.7 Å). The side chain of 4d was inserted into the catalytic channel of HDAC1, binding the cofactor Zn 2+ ion. The binding mode of 4d and HDAC3 is shown in Figure 3. The 4d-HDAC3 complex showed that a stronger inhibitor-enzyme interaction consists in two hydrogen bonds between 4d and Asp92 (2.1 Å) and Gly143 (3.3 Å) of HDAC3. The coordination of the benzotriazole group to the Zn 2+ ion (3.2 Å) can also be observed. Compound 5k showed a selectivity to HDAC2. The 5k-HDAC2 complex showed a complete insertion of its aromatic ring into the active site pocket, with multiple contacts with the tubular channel. The major interaction of 5k and HDAC2 ( Figure 4) consisted of three hydrogen bonds between 5k and Gly154 (2.7 Å), Gly143 (2.7 Å), Leu276 (3.1 Å), the π-π interaction with Phe210, as well as the coordination of the aromatic ring to the Zn 2+ ion (2.8 Å). The most potent compound in vitro 5j showed selectivity to HDAC1. The 5j-HDAC1 complex shows that the stronger inhibitor enzyme interaction consists of three hydrogen bonds between 5j and Phe150 (2.7 Å), Cys151 (2.7 Å), Gly300 (2.8 Å), the π-π interactions with Phe150, His140, and His178; as well as the coordination of the aromatic ring to the Zn 2+ ion (2.0 Å) ( Figure 5). The in vitro HDAC inhibitions of the obtained compounds were carried out with the assay kit containing mixed-HDAC isoforms. In order to further investigate the isoform selectivity of the four most potent HDAC inhibitors, an in silico experiment was performed with each HDAC isoform. Therefore, the results were not fully related. However, compounds with good binding affinities should have the potential to be specific HDAC inhibitors for each isoform.       The in vitro HDAC inhibitions of the obtained compounds were carried out with the assay kit containing mixed-HDAC isoforms. In order to further investigate the isoform selectivity of the four most potent HDAC inhibitors, an in silico experiment was performed with each HDAC isoform. Therefore, the results were not fully related. However, compounds with good binding affinities should have the potential to be specific HDAC inhibitors for each isoform.
To complete the evaluation of these potent HDAC inhibitors, antiproliferative activities were determined in human cervical cancer (HeLa), human colon cancer (HCT116), human breast adenocarcinoma cancer (MCF-7), and cholangiocarcinoma (KKU-100 and KKU-M213B) cells. The results, as shown in Tables 3 and 4, indicate that the derivatives 4c, 4d, 5j and 5k were less toxic to non-cancer cells than the lead compound. Compound 5j showed the best activity against HeLa, HCT116 and MCF-7, with IC50 values of 8.09, 9.65 and 11.57 µ M, respectively. Additionally, compound 5j displayed a strong antiproliferative activity against cholangiocarcinoma cells, as shown in Table 4. Compounds 4c and 4d exhibited antiproliferative activity against cholangiocarcinoma cells with selectivity towards KKU-100 and KKU-M213B, respectively. Compound 5k, the least toxic compound, showed a good activity against HeLa cells with an IC50 value of 23.14 µ M. Moreover, compound 5k appeared to be about 5-fold more selective towards HeLa cells than non-cancer cells, whereas [6]-shogaol (4) exhibited only a 3-fold selectivity. To complete the evaluation of these potent HDAC inhibitors, antiproliferative activities were determined in human cervical cancer (HeLa), human colon cancer (HCT116), human breast adenocarcinoma cancer (MCF-7), and cholangiocarcinoma (KKU-100 and KKU-M213B) cells. The results, as shown in Tables 3 and 4, indicate that the derivatives 4c, 4d, 5j and 5k were less toxic to non-cancer cells than the lead compound. Compound 5j showed the best activity against HeLa, HCT116 and MCF-7, with IC 50 values of 8.09, 9.65 and 11.57 µM, respectively. Additionally, compound 5j displayed a strong antiproliferative activity against cholangiocarcinoma cells, as shown in Table 4. Compounds 4c and 4d exhibited antiproliferative activity against cholangiocarcinoma cells with selectivity towards KKU-100 and KKU-M213B, respectively. Compound 5k, the least toxic compound, showed a good activity against HeLa cells with an IC 50 value of 23.14 µM. Moreover, compound 5k appeared to be about 5-fold more selective towards HeLa cells than non-cancer cells, whereas [6]-shogaol (4) exhibited only a 3-fold selectivity.
Anticancer activities of the best-four HDAC inhibitors were investigated upon the proliferation of five human cancer cell lines in a time-dependent manner. All selected compounds showed an antiproliferative activity in the growth inhibition of cancer cell lines. The results support the use of these HDAC inhibitors as potential anticancer candidates.

Conclusions
In conclusion, a series of new immino and pyrazole derivatives of [6]-shogaol were designed and synthesized as potential HDACs inhibitors. All derivatives exhibited HDAC inhibitory activity in the micromolar concentration ranges. Among these derivatives, pyrazole 5j with p-nitro substituent displayed the most remarkable HDACs inhibitory activities. Additionally, pyrazole 5j showed the best activity against all tested cancer cells among the synthesized derivatives and the most selective binding to HDAC1 based on molecular docking experiments. Therefore, the in vivo experiments of 5j will be further investigated.

General
Reagents were purchased from commercial sources (Sigma-Aldrich, Merck, and Carlo Erba). Reactions were monitored using analytical TLC plates (Merck, silica gel 60 F 254 ), and compounds were visualized under ultraviolet light. Silica gel grade 60 (230-400 mesh, Merck) was used for column chromatography. The NMR spectra were recorded in the indicated solvents on a Varian Mercury Plus spectrometer operated at 400 MHz ( 1 H) or 100 MHz ( 13 C). The IR spectra were obtained on Perkin Elmer Spectrum One FT-IR spectrophotometer. Mass spectra were determined using a Micromass Q-TOF 2 hybrid quadrupole time-of-flight (Q-TOF) mass spectrometer with a Z-spray ES source.

HDAC Activity Assay
The semi-synthetic derivatives were evaluated for their ability to inhibit HDAC enzymes. Inhibition of HDAC activity in vitro was assessed using the Fluor-de-Lys HDAC activity assay kit (Biomol, Enzo Life Sciences International, Inc., Farmingdale, NY, USA). The HeLa nuclear extract provided with the kit was used as a source of HDAC enzymes for the in vitro study. TSA was used as the positive control. The HeLa nuclear extract, substrate, buffer and inhibitors were incubated. Deacetylation of the substrate was performed next by adding a developer to generate a fluorophore. The spectra Max Gemini XPS microplate spectrofluorometer (Molecular Devices, San Jose, CA, USA) was used to measure fluorescence signal with excitation at 360 nm and emission at 460 nm. A decrease in fluorescence signal indicated an inhibition of HDAC activity. Trichostatin A (TSA) was used as a positive control. All experiments were carried out in triplicate.

MTT Assay
The MTT reduction assay was performed with non-cancer (Vero), human cervical cancer (HeLa), human colon cancer (HCT116), human breast adenocarcinoma cancer (MCF-7), and cholangiocarcinoma (H-69 (Vero), KKU-100 and KKU-M213B) cell lines according to the previously described method [34,35]. Briefly, cells were seeded in 96-well plates. The next day, cells were exposed to the selected compounds at various concentrations and incubated for 24, 48 and 72 h. After incubation, the culture medium was exchanged with 110 µL of MTT (0.5 mg/mL in PBS medium) and further incubated for 2 h. The amount of MTT formazan product was determined after dissolving in DMSO by measuring its absorbance with a microplate reader (Bio-Rad Laboratories, Hercules, CA, USA) at a test wavelength of 550 nm and a reference wavelength of 655 nm. The cell viability was expressed as a percentage of the viable cells of control culture conditions, and the IC 50 values of each group were calculated.

Molecular Docking Studies
The crystal structures of HDAC1, HDAC2, HDAC3 and HDAC8 (PDB entry code: 4BKX, 3MAX, 4A69 and 1T64, respectively) were obtained from the Protein Data Bank (http://www.rcsb.org/pdb (accessed on 1 March 2022). All water and non-interacting ions as well as ligands were removed. Then, all missing hydrogen and side-chain atoms were added using the ADT program. Gasteiger charges were calculated for the system. For ligand setup, the molecular modeling program Gaussview was used to build the ligands. These ligands were optimized with the AM1 level by using Gaussian03W. Molecular docking studies were performed for 50 runs using AutoDockTools 1.5.4 (ADT) (La Jolla, CA, USA) and AutoDock 4.2 programs and Lamarckian genetic algorithm search. A grid box size of 60 × 60 × 60 points with a spacing of 0.375 Å between the grid points was implemented and covered almost the entire HDAC protein surface. For TSA and other inhibitors, the single bonds were treated as active torsional bonds.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.