ADMET profiling and molecular docking of pyrazole and pyrazolines derivatives as antimicrobial agents

2.1 Biological data set

For the present study, twenty-three molecules of new pyrazoline and pyrazole derivatives were selected from the literature and were docked with the organism: Escherichia coli (Hassan 2013). Initially, the 2D structure of the compounds was designed using the ChemDraw professional 15.0 software, Chem3D 15.0 was then used to convert all compounds from 2D to 3D structures and they are optimized by this software (Ouabane et al. 2022). Finally, the molecules were saved in PDBQT file format by using Autodock tools. The structure of the studied molecules represented in Table 1.

Table 1 The structure of pyrazoline and pyrazol derivatives.

molecules	structures	molecules	Structure
M1		M13
M2		M14
M3		M15
M4		M16
M5		M17
M6		M18
M7		M19
M8		M20
M9		M21
M10		M22
M11		M23
M12

2.2 Molecular docking

The docking study contains three steps: the first step is Preparation/Optimization of the protein, the second step is Preparation of ligands and the last step is Molecular docking analysis (Belghalia et al. 2023).

In this work, we have focused on a bacterium that is found in the digestive tract of humans and warm-blooded animals and that causes several risks, this bacterium is Escherichia coli. The crystal structures of Escherichia coli (PDB ID: 1FJ4; Resolution: 2.35 Å), was obtained from protein data bank (PDB) (https://www.rcsb.org/pdb/home/home.do) (See Fig. 1. Moreover, the software AutoDockTools was used to adjust gasteiger charges, add polar hydrogen, and save the protein structure in the PDBQT file format (Flamandita et al., 2020). The enzymes are constituted of two chains; we chose one chain and delete the water molecules and ligand to allow easier use of the enzyme with a free active site during molecular docking (Abchir et al. 2022). The next step is the preparation of the grid, the grid plane should be in the center near the binding site, and it should be large enough to contain all the residues that interact with the ligand and large enough to put the ligand to have a full rotation, then the enzyme was placed in the center of the grid box while the number of grid points in the X, Y and Z directions was kept to the maximum (Belhassan et al. 2022). For the second step, the ligand library was collected from the literature, the 2D structures of the ligands were converted to.pdb format using autodock tools, Gasteiger charges were added, non-polar hydrogen was fused, and rotational bonds were detected and defined (Ouabane et al. 2023). Then, the docking was performed using autodock vina. The last step is devoted to the analysis of the docking results; we focus on the results of the simulation of the protein–ligand interaction that allowed us to obtain the best conformations with the binding energy, the conformation of the ligand with the protein that had a minimum energy present the best result for docking.

Close

Fig. 1

Co-crystal structure of protein 1FJ4.pdb.

Export to PPT

In this work, docking analysis was performed using the AutoDock Vina software. Using a number of points in the (x,y,z) = (20, 20, 20) Å box, Autodock Tools performed the GRID box, after docking we used Discovery Studio Visualizer for examined the structures and analyzed the interaction between receptor and the protein.

2.3 Prediction of ADME/Tox and bioavailability

Chemical database contain 23 molecules in the SMILES format are submitted one by one to each web server pkCSM provides information about Absorption, Distribution, Metabolism, Elimination, and Toxicity (ADMET). The most stable molecules that have the best interactions with the protein of Escherichia coli are also studied by calculating the Lipinski’s rule of five by using SwissADME considering molecular weight, log P, H-bond acceptors, H-bond donors and rotatable bonds, the Lipinski’s rule of five was used to evaluate the drug likeness of the potential inhibitor. Lipinski noted that molecules that comply with these 5 rules remain bioavailable (Stitou et al. 2021). According to the RO5, a drug-like compound should have a molecular weight (MW) < 500 g/mol, a log P < 5, hydrogen bond donors (HBDs) < 5, hydrogen bond acceptor (HBA) < 10 and rotatable bonds < 10 (En-nahli et al. 2022). A substance will have improved pharmacokinetic characteristics and greater bioavailability in the organism's metabolic process if it complies with the five principles (Ouassaf, Belaidi, Mogren Al Mogren, et al. 2021).

3.1 Docking results

In order to define theoretically the antibacterial mechanism of twenty-three Pyrazoline and Pyrazole Derivatives, a molecular docking study was carried out using AUTODOCK Vina. All the study compounds were docked into the crystal structure of Escherichia coli (PDB ID: 1FJ4) with a resolution of 2.35 Å. The first step in docking protocol is validation by redocking the reference ligand. Indeed, it was found that such ligand was attached to the binding site of the bacteria, through the same interactions i.e H-bond with His 298 and His 333 as shown in Fig. 2.

Close

Fig. 2

3D view Superposition of the reference ligand on the 1FJ4 pocket (the red stick represents the reference ligand; the green stick represents the redocked ligand).

Export to PPT

The results shown in Table 2 indicate that compounds M6, M17, M19 and M20 are stabilized in 1FJ4 receptor pocket by various interactions with very low binding affinities of −9.6, −9.3, −9.5 and −10.3 (Kcal/mol) respectively. And the predicted docking results of these four compounds are shown in Fig. 3. We noted that a combination of hydrogen bonding and hydrophobic interaction maintained these three molecules in the active pocket. The interaction of the hydrogen bond was observed with the amino acid residues Thr (3 0 2), Thr (3 0 0), Val (2 7 0), and His (A: 298). And the major hydrophobic contacts between protein and compounds Ala (2 7 1), Pro (3 0 3), Ala (2 0 6), Pro (2 7 2), Phe (2 2 9), Cys (1 6 3), Val (3 0 4), His (3 3 3), Phe (3 9 2), His (2 9 8). These findings demonstrated the significance of hydrophobic interactions and hydrogen bonds in the antibacterial activity of pyrazoline and pyrazolone derivatives.

Close

Fig. 3

2D view of the binding conformations and hydrogen bond interactions of the four inhibitors at the active site of 1FJ4.

Export to PPT

3.2 Lipinski’s rule and ADMET prediction

Absorption, distribution, metabolism, excretion, and toxicity (ADMET) play key roles in drug discovery. In our study, prediction of the properties of ADMET was carried out using pKCSM web server; Prediction of ADMET parameters is listed in Table 3. This method is very important to select best molecules we can use as a drug candidate.

We are interested in M6, M17, M19 and M20 the most potent inhibitor in the dataset. As a result, we can observe that these compounds have high calculated values of intestinal absorption by humans, which are larger than 80%., compared to the criterion, an intestinal absorption value of<30% indicates poor absorbance. This indicates that these molecules can be easily absorbed by the intestine and circulate in the bloodstream (Norinder and Bergström 2006).

The distribution analysis shows that the molecules M19 and M20 are poorly distributed in the brain with values less than −1 (Abdelrheem et al. 2021), on the other hand, the molecules M6 and M17 present values higher than −1 which indicates that these two molecules are easily distributed in the brain.

Measures such as the permeability of the blood-surface area of the central nervous system (CNS) permeability (log PS) can be obtained from in situ brain perfusions with a direct injection of the compound into the carotid artery, which create a lakes of systemic distribution effects that could distort brain penetration. In this case compounds that present log PS > -2 are able to penetrate the CNS. However, compounds with log PS < -3 are unable to penetrate the CNS.

From Table 3 it can be seen that the compound M19 is unable to penetrate the CNS, while the other (M6, M17, M20) are capable to penetrate the CNS (Larik et al., 2019). In addition, metabolizing enzymes is the major study in phase I in discovery of drug, as indicate in Table 3, cytochrome P450 (CYP) includes substrate and inhibitor enzyme, the most important cytochromes P450 are CYP 2D6 and CYP 3A4 and it is involved in the metabolism of approximately half of the drugs currently in use, the results obtained indicate that M6, M17, M19 and M20 are non-substrate and non-inhibitor of 2D6 CYP enzyme. On the other hand, they are substrate and inhibitor of 3A4 CYP except for M17 is no substrate of the last enzyme (Moroy et al. 2012).

The Lipinski’s rule including molecular weight, number of hydrogen bonds donor, number hydrogen bonds acceptor, number of rotatable bonds and log P were shown in Table 4.

In the filter that combines the drug-likeness criteria of Lipinski's rule of five (MW ≤ 500, log P ≤ 5, H-bond donors ≤ 5 and H-bond acceptors ≤ 10) and rotatable bonds ≤ 10) (da Cunha Xavier et al. 2021). Lipinski's rule of five is strictly adhered to by the four compounds, which indicates that they exhibit strong drug-like properties. For excretion parameters, the molecule M19 is considered an OCT2 renal substrate. For toxicity parameters, the AMES test results suggest that only molecule M11 is active as AMES toxicity, but the four molecules’ studies are also predicted to be hepatotoxic.