Thiazole–Chalcone Hybrids as Prospective Antitubercular and Antiproliferative Agents: Design, Synthesis, Biological, Molecular Docking Studies and In Silico ADME Evaluation

Compounds bearing thiazole and chalcone pharmacophores have been reported to possess excellent antitubercular and anticancer activities. In view of this, we designed, synthesized and characterized a novel series of thiazole–chalcone hybrids (1–20) and further evaluated them for antitubercular and antiproliferative activities by employing standard protocols. Among the twenty compounds, chalcones 12 and 7, containing 2,4-difluorophenyl and 2,4-dichlorophenyl groups, showed potential antitubercular activity higher than the standard pyrazinamide (MIC = 25.34 µM) with MICs of 2.43 and 4.41 µM, respectively. Chalcone 20 containing heteroaryl 2-thiazolyl moiety exhibited promising antiproliferative activity against the prostate cancer cell line (DU-145), higher than the standard methotrexate (IC50 = 11 ± 1 µM) with an IC50 value of 6.86 ± 1 µM. Furthermore, cytotoxicity studies of these compounds against normal human liver cell lines (L02) revealed that the target molecules were comparatively less selective against L02. Additional computational studies using AutoDock predicted the key binding interactions responsible for the activity and the SwissADME tool computed the in silico drug likeliness properties. The lead compounds generated through this study, create a way for the optimization and development of novel drugs against tuberculosis infections and prostate cancer.


Chemistry
The target thiazole-chalcone hybrids were synthesized by the condensing of different substituted aromatic ketones with 2,4-dichlorothiazole-5-carboxaldehyde in the presence of a glacial acetic acid and hydrochloric acid mixture to isolate the compounds (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) in 75-91% yields. The synthesized hybrids were pale yellow-colored compounds with solubility in chloroform, methanol and DMSO. All the compounds were characterized by FT-IR, 1 H-NMR and mass spectrometry and the most potent compounds 7, 12 and 20 were also characterized by 13 C-NMR spectroscopy. In the FT-IR spectrum, two diagnostic stretching absorption bands of C=O and HC=CH were seen around the wave numbers 1651-1698 cm −1 and 1506-1520 cm −1 , respectively. The 1 H-NMR spectra showed two doublet signals characteristic of the αand β-protons of the propenone linkage resonating between the chemical shift values 7.27-7.89 ppm and 7.66-8.16 ppm. The coupling constant value J, for these doublets ranged between 15-17 Hz. These large coupling constant values confirmed the trans isomer of the olefinic bond present in the hybrids. 13  the formation of chalcones. Additionally, all the compounds showed an isotopic M + 2 peak of one-third intensity to the molecular ion peak. The FT-IR, 1 H NMR, 13 C NMR and Mass Spectra for compounds can be found in the Supplementary Materials.
In the monosubstituted series, the ortho (1, 9) and para (3,9) positions were found to be beneficial for activity over the meta (2, 10) position Among the monosubstituted chalcones 1-3, 9-11, the ortho and para-F substituted compounds showed 3.79-fold better activity over the Cl-containing compounds (MIC 20.68 ± 2 µM (9, 11) vs. 78.46 ± 1 µM (1, 3)). The MIC value was further improved when the phenyl ring was disubstituted with F over Cl  (7)). Compounds 12 and 7 bearing halogen atoms in both ortho and para positions, i.e., 2,4-difluorophenyl (MIC = 2.43 ± 1 µM) and 2,4dichlorophenyl (MIC = 4.41 ± 2 µM) showed potencies 10.42 and 5.74 times more than the standard (MIC 25.34 ± 2 µM) whereas the monofluorinated compounds 9 and 11 bearing fluorine atoms at ortho and para positions, showed activity at MIC 20.68 µM which was 0.81 times greater than pyrazinamide (25.34 ± 2 µM). The chalcone 14 (MIC = 9.74 ± 2 µM) containing 2,6-difluorophenyl scaffold was 2.6-times more active than the standard. This suggests that the degree of electronegativity played a key role in modulating the physicochemical properties of the phenyl ring thereby influencing the antitubercular activity. Among the bioisosteres 17-20, the activity ranged from 343.45 ± 1 to 701.40 ± 1 µM. No improvement in activity was observed over the standard pyrazinamide. b Data presented as mean ± SD (n = 3). All the compounds and the standard dissolved in DMSO, diluted with culture medium containing 0.1% DMSO. The control cells were treated with culture medium containing 0.1% DMSO. Bold numerical value in third column represents compound with highest activity.
A summary of the structure-activity relationships (SAR) of novel thiazole-chalcone hybrids and the most potent antitubercular and antiproliferative compounds retrieved from this study are depicted in Figure 3.

Molecular Docking Studies
Molecular docking was carried out for selected compounds against the potential antitubercular and anticancer targets-Isocitrate Lyase and Topoisomerase IIa ATPase, respectively. The in silico antitubercular and anticancer activity results of selected ligands against Isocitrate Lyase and Topoisomerase IIa ATPase were reported in terms of binding energy and ligand interactions with amino acid residues at active pocket of proteins. The in silico antitubercular results indicated that all compounds (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) showed strong binding affinity (ranges from −5.7 to −7.3, given in Table 3) towards the amino acid residues in active pocket Isocitrate Lyase protein through H-bond and hydrophobic interactions, compared to standard drug Pyrazinamide. The selected compounds having the electron withdrawing group substitution on the aromatic ring in place of the amine group may be the reason for the high affinity of these compounds compared to the standard drug pyrazinamide. Compounds 2 and 16 showed the highest binding affinity (−7.3). Table 3. Docking scores of thiazole-chalcone hybrids against Isocitrate Lyase protein and Topoisomerase IIa ATPase.

Compound
Binding Affinity Compound Binding Affinity Among all the compounds, compound 2 interacted with Trp320 amino acid residue through the H-bond and with Leu69, Cys314, Phe332, Ile346, Ala349, His352 amino acid residues through hydrophobic interaction. Similarly compounds 16, 15, 5, 6, 12 and other compounds also showed H-bond interactions with Trp320 and hydrophobic interaction but the standard drug pyrazinamide has H-bond interaction with Asp153, Arg228, Glu285, Asp108, Ser91, Leu348 amino acid residues and has hydrophobic interaction with Cys191, Thr347, His180 amino acid residues (Figures 4 and 5; Table 4). Moreover, the hydrophobic interactions between the functional groups of compounds and amino acid residues are more, compared to the hydrophobic interactions between the functional groups of standard drug pyrazinamide functional groups and amino acid residues of target protein. The compounds having the electron withdrawing group substituted on the aromatic ring in place of the amine group, and compounds having H-bond interaction with Trp320 amino acid residues and a greater number of hydrophobic interactions present between compounds and amino acid residues in the active site of the target protein may be the possible reason for the high affinity of the compounds when compared to standard drug pyrazinamide.

In Silico Drug Likeliness Studies
Some selected compounds 7, 14 and 20 which showed the best activity in antitubercular and antiproliferative activity were computed for certain properties using

In Silico Drug Likeliness Studies
Some selected compounds 7, 14 and 20 which showed the best activity in antitubercular and antiproliferative activity were computed for certain properties using web-based SwissADME software (Table 6). It can be observed that they inhibited CYP2C19, but did not inhibit CYP2D6. However, they showed high GI absorption and passed the Lipinski Rule of five. Hence, these molecules have good drug-like properties and they can be taken as leads for further in vivo investigation.

Chemistry
General protocol for the synthesis of thiazole-chalcone hybrids (1-20): Initially, 1 mmol of 2,4-dichlorothiazole-5-carboxaldehyde was dissolved in a mixture of glacial acetic acid (4 mL) and concentrated hydrochloric acid (2 mL) [60]. To the above solution, 1 mmol of corresponding aromatic ketone dissolved in 10 mL of ethanol was transferred and refluxed for 4-6 h. After completion of the reaction, the precipitate of the target compound was separated by filtration and washed thoroughly with cold water (50 mL × 2) and dried in a desiccator (Scheme 1). The crude precipitate was purified by column chromatography on silica gel using a mixture of hexane and ethyl acetate (2-25%).

Antitubercular Activity
The antitubercular activity of all the target compounds (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) was tested against the Mycobacterium tuberculosis H37Rv strain. Pyrazinamide was used as the reference standard. In the present investigation, we employed the protocol described in the literature [61,62]. The icy culture of Mycobacterium tuberculosis H37Rv strain in Middlebrook 7H9 broth with the addition of 0.2% glycerol and 10% albumin-dextrose-catalase was defrosted and diluted in broth to 10 5 CFU mL −1 (colony forming unit/mL) dilutions. Separately, test compounds were dissolved in Dimethyl sulfoxide (DMSO) and later diluted with broth to achieve a concentration that was two-times the required concentration. Throughout this experiment, the final concentration of DMSO was 1.3% in the assay. Each test-tube used was then inoculated with 0.05 mL of standardized culture and then incubated for 21 days at 37 • C. The growth in the test-tubes was compared with the positive control, pyrazinamide and negative control, i.e., without inoculum and the drug. The minimum inhibitory concentration (MIC) of the individual target compound was determined by broth dilution assay. The MIC values were obtained in µg/mL and further these values were converted to micromoles (µM) considering the structural diversity of standard drug and the target compounds to draw a more meaningful conclusion.

Antiproliferative and Cytotoxic Activity
The in vitro antiproliferative and cytotoxicity of compounds 1-20 was evaluated by Mosmann's MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay method, as described in the literature [63][64][65] on prostate cancer cell lines (DU-145). In both the assays, the IC 50 values of the tested compounds were compared with the positive control Methotrexate (Mtx). The reduction of the soluble MTT to blue-color formazan is the principle underlined in MTT assay and such change is chiefly due to the action of intracellular mitochondrial reductase in the living cells. The prostate cancer cell lines (DU-145), were cultured in Dulbecco's Modified Eagle Medium (DMEM) media at 37 • C and humidified at 5% CO 2 was used to culture prostate cancer cell lines (DU-145). The compounds (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) were initially dissolved in 0.1% DMSO to prepare their stock solutions. Later, the desired concentrations of the compounds were achieved by dissolving in sterile water. The cells were transferred on to 96-well plates at 100 µL total volume and with a density of 1 × 10 4 cells per well. The cells were permitted to adhere for a period of 24 h and then the assay medium was replaced with a fresh medium containing target compounds and incubated in DMEM with 10% fetal bovine serum (FBS) medium at 37 • C for an additional 48 h. Later, the medium was replaced with 90 µL of fresh DMEM without FBS. The above wells were treated with 10 µL of MTT reagent (5 mg/mL of stock solution in DMEM without FBS) and incubated for 3-4 h at 37 • C. The formed blue formazan crystals were dissolved in 200 µL of DMSO. The optical density was then determined at 570 nm using a micro plate reader. The assay was executed in triplicate for three independent experiments. The same trialing was also performed to confirm that the negative control-DMSO had no effect in the study. The results had good reproducibility between replicate wells with standard errors below 10%. The IC 50 values measured in µg/mL for the antiproliferative were converted and expressed in µM. Additionally, all the compounds were also assessed for their cytotoxic activity on the normal human liver cell lines (L02) using the same protocol discussed above.

Molecular Docking Studies
The X-ray crystal structures of Isocitrate Lyase (PDB ID: 1F8M) and Topoisomerase IIa ATPase (PDB ID: 1ZXM) were taken from protein data bank (rcsb.com/pdbdatabase (accessed on 7 April 2021)). Using PyMOL 2.3.4 the water molecules were removed and hydrogens were added as well as co-crystal ligands were extracted and saved in mol2 format. The mol2 format file of protein was loaded and then converted to pdbqt format using Autodock module Macromolecule tool in PyRx Virtual screening software 0.8 [66]. The 2D-structures of the target compounds (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) and that of the standard drugspyrazinamide and methotrexate were drawn in ChemDraw ultra 12.0 and saved as sdf file. The ligand files were subjected to energy minimization (force field-uff) through Open babel tool and then conformers for the selected ligands were generated through AutoDock pdbqt files in PyRx Virtual screening software 0.8. The docking was then performed through PyRx Virtual screening software 0.8 combined with AutoDock Vina, Open babel, Python shell tools. The prepared protein file and ligand files were selected through Vina module and the grid box was selected according to the previously reported amino acid residues by adjusting the x, y, z coordinates of grid box, then run the Vina. The results were analyzed by using DS visualizer software to visualize the interactions between ligands and amino acid residues of active site of protein (Kwofie et al., 2018).

In Silico Drug Likeliness Studies
To meet the requirements of the drug-likeliness, the properties of the most potent compounds 7, 14 and 20 target compounds were evaluated for their in silico parameters including GI absorption, Lipinski rule of five as well as CYP2C19 CYP2D6 inhibition using SwissADME web (http://www.swissadme.ch/ (accessed on 24 March 2021)) [67].

Conclusions
In the present study we designed and synthesized 20 new thiazole-chalcone hybrids and tested all the compounds for antitubercular, antiproliferative and cytotoxic activities against Mycobacterium tuberculosis H37Rv strain, DU-145 (prostate cancer cell line) and human liver normal cells-LO2, respectively. We identified five potential antitubercular chalcones and one promising antiproliferative chalcone as they displayed better activity than the standard drugs. Among the five potent hybrids, the compounds 12 and 7 exhibited the strongest antitubercular activity with MIC values 2.43 and 4.41 µM correspondingly. The top most active antiproliferative hybrid 20 showed activity at IC 50 value 6.86 ± 1 µM. All the compounds were subjected to molecular docking studies against antitubercular and anticancer drug targets and a good correlation was observed between the in vitro and docking results. Additionally, the calculated SwissADME properties of the most potent compounds 7, 14 and 20 were in agreement with the required drug-like properties. All these results clearly confirm the usefulness of the active compounds for the furtherance of drug discovery and development against tuberculosis and prostate cancer.