Bis-triazoles MS44-53 were prepared as previously described [16].

Human melanoma MDA-MB-435, pancreatic carcinoma MIA PaCa-2, colorectal carcinoma HCT-116, colorectal adenocarcinoma COLO 205, and skin normal fibroblast CCD-1064Sk cell lines were purchased from American Type Culture Collection (Manassas, VA, USA) (ATCC), USA and stored in liquid nitrogen. Cells were passaged twice weekly upon reaching 70–80% confluency, and low passage numbers (≤ 20) were used in all experiments. Lonza MycoAlert™ mycoplasma detection kits were used (according to the manufacturer's instructions), ensuring cultures were contamination-free. Cells were sub-cultured in RPMI 1640 medium (Sigma-Aldrich (UK), CN: R8758) supplemented with 10% (v/v) heat-inactivated fetal bovine serum (Gibco, CA, USA), 100 units/mL penicillin, and 100 μg/mL of streptomycin (Gibco 15140-122). Cells were maintained at 37 °C in a humidified 5% CO₂ incubator (Shel Lab., Cornelius, OR, USA).

The method has been described previously by Saleh et al. [16] and was adapted from Mosmann [18]. MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] reagent was used to evaluate the antiproliferative activity of ligands MS47 and MS49 on the human melanoma MDA-MB-435, pancreatic carcinoma MIA PaCa-2, colorectal carcinoma HCT-116, colorectal adenocarcinoma COLO 205, and skin normal fibroblast CCD-1064Sk cell lines. Cells were seeded into 96-well plates (Nunc, Roskilde, Denmark) at the density of 3 × 10³ per well (180 µL per well) and allowed to adhere for 24 h at 37 °C/5% CO₂. Agent top stock solutions (10 mM in DMSO) were then freshly made. Serial dilutions were prepared in RPMI 1640 medium for addition to carcinoma cells. Control wells received vehicle alone (20 µL per well). Final test agent concentrations in the wells were; 0.01, 0.05, 0.1, 0.5, 1, 5, 10, 50, and 100 µM. The final concentration of DMSO in the wells never exceeded 1%. Vehicle control assays were performed (0.0001–1% DMSO). Experimental plates were incubated for a further 72 h period at 37 °C/5% CO₂. Cell viability was recorded at the time of agent addition (T₀ ) and after 72 h exposure: following the addition of MTT solution (2 mg mL^-1 in PBS: 50 µL per well), experimental plates were incubated for 3 h to allow reduction of MTT by viable cells to insoluble dark purple formazan crystals. The supernatant in each well was then aspirated, and the cellular formazan was solubilized by the addition of DMSO (150 µL per well). Absorbance (OD) was read at a wavelength of 550 nm by using an Anthos Labtec system plate reader. The measured intensity is proportional to metabolic activity, which correlates with viable cell number. Estimated IC₅₀ values (test agent concentrations that inhibit cell growth by 50%) were calculated using Prism (version 7.0) software. Results are expressed as the mean of three independent experiments (n = 8 per trial).

The method applied for this study was adapted from Schuster [20]. Bis-triazoles MS44-53 were profiled against an in-house established pharmacophore database. The database consists of a collection of 106 pharmacophore models. The pharmacophore model collection (PMC) reflects different binding modes of 34 drug targets.

The PMC was established in the Molecular Modeling Research Laboratory at the University of Jordan, and the pharmacophores were created mainly by the application of the DiscoveryStudio^® (version 4.5) software suite from Accelrys Inc. (www.accelrys.com). Each one of the pharmacophore models has been evaluated individually for its predictive power, using the receiver operating characteristic - area under the curve (ROC-AUC), and all have an AUC ≥ 0.8. The ligands were profiled using the automated Ligand Profiler protocol in the DiscoveryStudio® (version 4.5) software, and mapped to all 106 pharmacophores in the PMC database. The output of profiling includes fit values (0 to 1) of each molecule profiled against each respective pharmacophore. The expressed fit value predicts the activity profiles of ligands. The higher fit value is indicative of the bioactivity of the ligand against a drug target represented by its respective pharmacophore(s) model.

The PMC consists of both ligand based and structure based pharmacophores. Ligand based models were constructed by following a systematic method that includes the following steps: Collecting ligands known to be active against a specified target and grouping them into training sets. Exploring the pharmacophoric space of the training sets and automatically generating pharmacophore models with the application of CATALYST-HYPOGEN module of DiscoveryStudio^®. The generated pharmacophores undergo QSAR analysis to nominate the best possible pharmacophore models capable of explaining bioactivity of the training sets. The models’ quality is evaluated using receiver operating characteristic-area under the curve (ROC-AUC). The validation step defines the pharmacophore model’s ability to successfully classify a long list of external compounds (known actives and structurally similar decoys) as actives or inactives. Subsequent addition of steric exclusion spheres is applied using HIPHOP-REFINE module of DiscoveryStudio^® to define the steric constraints of the binding pocket and to enhance the quality of the models [21-25].

As for structure-based pharmacophore models, the method starts with the collection of known bioactive molecules against a certain target. These molecules are docked in the protein structure obtained from the Protein Data Bank, and then the docked poses are scored. Docking and scoring are performed using the different docking and scoring engines found in DiscoveryStudio^® software. After verifying the best docking configuration, Docking-based comparative intermolecular contacts analysis (dbCICA) is applied. Pharmocophore models are then manually built according to the critical points of ligand-protein interaction that were determined by dbCICA. The models’ quality is validated using ROC-AUC and exclusion spheres are added, just like in the ligand-based pharmacophore models [21, 22, 26-28].

The antiproliferative activities of MS47 and MS49 were evaluated in vitro using the MTT anti-proliferative assay [18, 29] against MDA-MB-435 (human carcinoma), MIA PaCa-2 (pancreatic carcinoma), HCT-116 (colorectal carcinoma), and COLO 205 (colorectal adenocarcinoma) cell lines. In addition, selective cytotoxicity for MS47 and MS49 was investigated utilizing CCD-1064Sk (human skin normal fibroblast) cell line. Estimated concentrations at which cell proliferation is inhibited by 50% (IC₅₀) were calculated from dose-response curves after 72 h exposure of the cells to MS47 and MS49. The IC₅₀ values are presented in Table 1, while dose-response curves are shown in Fig. (2>).

MS47 and MS49 show potent proliferation inhibitory activities against human carcinoma and normal fibroblast cell lines in the concentration range (0.01-100 µM) used. However, MS49 exhibits more potent proliferation inhibition activity than MS47 in human melanoma MDA-MB-435, pancreatic cancer MIA PaCa-2, colon cancer HCT-116 and COLO 205 cell lines. Notably, melanoma MDA-MB-435 cell line is the most sensitive cell line to ligand MS49, scoring the smallest nanomolar IC₅₀ value of 115.2 nM.

To investigate the safety profile of MS47 and MS49, their cytotoxic effect against non-cancerous cells was assayed with the use of CCD-1064Sk (human skin normal fibroblast) cell line. Ligand MS47 exhibits a higher IC₅₀ value against non-cancerous cells in comparison to its IC₅₀ values against melanoma MDA-MB-435, colon cancer HCT-116 and COLO 205 cell lines, with cancer selectivity indexes (SI) of 2.3, 1.7 and 2.4, respectively. While ligand MS49 shows a higher IC₅₀ value against skin normal fibroblasts in comparison to all tested human carcinoma cell lines; MDA-MB-435, MIA PaCa-2, HCT-116 and COLO 205, with SI of 7.6, 2.5, 4.5 and 4.5, respectively.

The results demonstrate that both ligands are potent cytotoxic compounds, however, MS49 has higher potency than MS47 against human carcinoma cells and less toxicity (better safety margin) in human normal cells. Consequently, further biological activity evaluations should be performed for ligand MS49.

Pharmacophore is the spatial arrangement of features essential for a molecule to interact with a specific molecular target. Pharmacophore profiling is a method that can be used as an alternative to molecular docking in order to screen a ligand and find if it is capable of interacting with a potential target [30].

Fig. (2). Dose-response curves that show the growth inhibitory effects of MS47 & MS49 against A) human melanoma MDA-MB-435, B) pancreatic carcinoma MIA PaCa-2, C) colorectal carcinoma HCT-116, D) colorectal adenocarcinoma COLO 205, and E) human skin normal CCD-1064Sk fibroblast cell lines. Values are mean ± SD, n = 8, graphs representative of experiments performed on at least three separate occasions.

Bis-triazoles MS44-53 were screened by pharmacophore-based activity profiling screen [20, 31], using Accelrys DiscoveryStudio^® 4.5 software, in order to predict their affinity to biological targets. The profiling screen presents the results as percentages of affinity of ligands to a target protein and reflects the probability of their inhibitory or induction capabilities towards these targets. The ligand structure with % binding probability ≥ 50% is believed to fit the binding pocket of the biological target and thus be very close to the proper pharmacophore. The % probability of binding of bis-triazoles MS44-53 to different biological targets; FGFR1, PDGFR2, FLT3, mTOR, PPAR-gamma, MUR-F and CETP, are shown in Table 2.

The results demonstrate noteworthy percentages of binding probability for MS44-53 towards the kinases involved in many cancer diseases, FGFR1, PDGFR2, FLT3, and mTOR, with ranges of 46-71%, 64-66%, 56-77% and 59-61%, respectively, making them promising novel anticancer drugs for the treatment of various carcinoma diseases. MS44-53 showed an interesting % binding probability (71-72%) to the nuclear receptor PPAR-gamma that increases insulin sensitization and enhances glucose metabolism, making them potential agonist hypoglycaemic drugs for treating diabetes type II disease. An antibacterial activity is predicted from the interesting % inhibition probability values (63%-66%) shown by our ligands to MUR-F, the key enzyme in bacterial cell wall biosynthesis in both gram-positive and gram-negative bacteria, making them potential antibacterial agents. Furthermore, the data show a significant % inhibition probability (47% to 77%) for MS44-53 to the plasma lipid transfer protein CETP that facilitates the transport of cholesteryl esters and triglycerides between lipoproteins, making the ligands potential drugs for treating cardiovascular diseases caused by LDL cholesterol.