0% READ
FULL TEXT
INTRODUCTION
The problem of Antimicrobial Resistance (AMR) has become a sharp public health issue on the global agenda and has already created an emergency need to continue searching and developing new, safer, and more effective antimicrobial agents to address it on a long-term basis (Salam et al., 2023). The growing rate of resistance in microbial strains has seriously reduced the clinical efficiencies of most of the remaining antibiotics, hence the need to adopt other forms of therapeutic interventions. Benzimidazole-based scaffolds have garnered significant attention in medicinal chemistry because of their pleiotropic and well-documented biological activities, such as antimicrobial, anthelmintic, antiviral, and anticancer activity (Monga et al., 2024; Chakrabarti, 2023). The structural versatility of the benzimidazole nucleus that allows extensive chemical modification is the main reason why this pharmacological profile is extensive. This type of structural flexibility allows adjustment of physicochemical properties, selectivity on targets, and pharmacokinetic behaviour, which eventually results in increased antimicrobial potency (Chakrabarti, 2023). Due to the heteroatoms and aromatic rings, benzimidazole analogs may successfully interact with the microbial target via hydrogen bonding, pi-pi stacking, and hydrophobic interactions. In parallel, another group of biologically active organic compounds is represented by Mannich bases. They are typified by the existence of a β amino carbonyl group and have been recorded to have varying pharmacological actions especially strong antimicrobial effects (Raju et al., 2023; Raoof and Sadiq, 2022; Yamali et al., 2023). The addition of the functionality of Mannich base has been reported to enhance the molecular lipophilicity, solubility, and membrane permeability, which are important factors of antimicrobial activity. Thus, the conjugation of the benzimidazole core with the moiety of a Mannich base seems to be a reasonable and promising solution to the design of the new antimicrobial agents. This hybridisation will likely be able to combine the favourable attributes of the two pharmacophores, which may result in synergy, increased activity on resistant microbial strains of S. aureus, and enhanced safety in combination with improved safety profiles (Wang et al., 2023; Zalaru et al., 2022). Past research studies on N-Mannich base derivatives of benzimidazole have shown great antibacterial, antifungal, and antiprotozoal effects, thus supporting the therapeutic promise of such hybrid molecules (Ablo et al., 2023). Computational methodologies have now become essential tools to the accelerated identification and optimisation of potential drug candidates in modern drug discovery, significantly saving time and cost compared to traditional experimental methods, which is what is referred to as the first and third way of drug discovery (Chakrabarti, 2023; Salam et al., 2023). Molecular docking is one of the most popular in silico methods, which allows predicting the binding orientation, interaction patterns, and binding affinities of small molecules in the open sites of biological targets, including essential microbial enzymes (Monga et al., 2024; Raju et al., 2023; Pattapulavar et al., 2025). Molecular docking is useful to understand the interaction between ligands and targets and help in the prioritisation of useful compounds that should be validated further through experiments. Besides docking studies, early ADMET property assessment is also of paramount importance in drug development. Exhibition of poor pharmacokinetic behaviour or toxicity profile are significant reasons of drug attrition in clinical trials (Salam et al., 2023; Monga et al., 2024; Chakrabarti, 2023). As a result, prediction of ADMET during drug design has become obligatory in the early stages of drug design as a method of determining drug-like candidates with favourable safety and pharmacokinetic profiles. In the current study, the computer-aided drug design technique was used to design and screen Mannich base-substituted benzimidazole analogs as potential antimicrobial agents. The interaction between the designed compounds and critical microbial target proteins were evaluated by molecular docking, and the pharmacokinetic characteristics as well as toxicity properties of these compounds were predicted by full-scale ADMET analysis simultaneously. The overall goal of the work is to find prospective lead candidates, which can be further developed into effective antimicrobial agents to deal with the rising challenge of antimicrobial resistance.
MATERIALS AND METHODS
Molecular Docking Studies
Virtual screening software was used to analyse the expected optimal binding conformation of projected ligands into the enzyme-binding pocket. The docking protocol was applied to testing of the binding affinities with the scoring functions accounting for both steric and electrostatic interactions allowing thereby screening a range of compounds for future lead optimisation. All the molecular modelling studies were performed by using Auto Dock 4.2 (Hryhoriv et al., 2021; Kingsley and Abraham, 2022; Edache et al., 2022; Srivastava, 2021). Lamarckian Genetic Algorithm (LGA) was utilised to examine the ligand conformational space and minimise the Binding affinity between the ligands and the target receptor. Auto Dock Tools was used to prepare the receptor and ligand structure and grid maps of the active site were created to define the docking search space. The docking simulations were carried out, and the results analysed according to the binding energy and the predicted interaction patterns.
Xray Crystal Structure
X-ray crystal structure of the target enzyme DNA gyrase (1KZN) Figure 1 are obtained from Protein Data Bank. For the docking study a crystal structure with 2.00 Å resolution was used.
Protein Preparation
The protein configuration were prepared by voiding water molecules, co-crystallisation ligands and any other non-protein constituents (Pattapulavar et al., 2025; Hryhoriv et al., 2021; Kingsley and Abraham, 2022). Polar hydrogens were added and Gasteiger partial charges were attributed to all atoms with use of Auto Dock Tools (Kingsley and Abraham, 2022; Edache et al., 2022). Afterwards, processed protein structures were retained in PDBQT format for extended molecular docking (Chakrabarti, 2023).
Ligand Preparation
ChemSketch was employed to rendered the two-dimensional configuration of the ten modeled benzimidazole derivatives. Figure 2 ligands with one substituent (Mannich base) and the reference drug ciprofloxacin. The 2D structures were then transformed to 3D structures and stored in PDB format through Open Babel. The 3D structures were then energy-minimised with the help of appropriate force fields to get the low-energy conformers which were stable (Chand et al., 2025; Edache et al., 2022). For molecular docking simulations, polar hydrogen was added and Gasteiger partial charge was assigned to all of the ligand molecules (Kingsley and Abraham, 2022; Edache et al., 2022). Finally, the ligands were optimized and retained in PDBQT format to perform docking simulations with AutoDock (Chakrabarti, 2023).
Molecular Docking
The simulations of molecular docking were conducted with AutoDock 4.2 (Salam et al., 2023; Chakrabarti, 2023). The target protein active sites were defined as a grid box that defines the search space of ligand binding. The size and position of the grid box was established either using co-crystallised ligands or using standard active site residues that have been reported in the previous studies (Pattapulavar et al., 2025; Chand et al., 2025; Kingsley and Abraham, 2022). The docking simulations were used to forecast the manner in which the ligands oriented themselves in the protein active site and calculate the respective binding energies. All the posed docked structures and binding energies were analysed and main interaction, such as hydrogen bonds, hydrophobic contacts, and pi-pi stacking Interactions were observed and visually represented (Edache et al., 2022).
ADMET Property Prediction
Computational in silico tools were used to evaluate the pharmacokinetic behaviour and toxicological risks of the designed ligands were:
Swiss ADME
This online platform was employed to forecast physicochemical characteristics, such as molecular weight, log P, Topological Polar Surface Area (TPSA), and water solubility. It also gave data on the pharmacokinetic properties including gastrointestinal absorption and penetration of the blood-brain barrier Figure 3. Also, the Lipinski, Ghose, Veber, Egan, and Muegge filters were used to determine drug-likeness, and the medicinal chemistry suitability of each drug was determined (Chakrabarti, 2023; Chanda and Katari, 2025; Tahiroğlu et al., 2025).
OSIRIS Property Explorer
The software was employed to compute the drug-likeness, drug scores and predict possible toxicity risks, such as mutagenicity, tumorigenicity, irritancy and reproductive toxicity. The findings were presented in the form of colour codes; green was low or no risk, orange medium risk and red high risk (Bhat et al., 2025; Salihu and Salleh, 2023; Shehab et al., 2024; Srivastava, 2021).
Molinspiration Cheminformatics
This programme was applicable to compute molecular properties including Log P, TPSA and the rotatable number of bonds, and to assess the adherence to Lipinski Rule of Five, giving an idea into the oral bioavailability and drug-likeness (Bhat et al., 2025; Salihu and Salleh, 2023; Panigrahi and Sahu, 2025). The molecules with a molecular weight less than 500 Dalton, less than 10 H-bond acceptors, less than 5 H-bond donors, and Log P less than 5 were deemed prone to have good oral bioavailability (Chand et al., 2025; Sampat et al., 2022).
Analysis and Visualization of Data
All the computational data such as docking data, docking results, ligand-protein interactions data and ADMET predictions were collected, arranged and reviewed. The correct statistical procedures were used, where required, to compare the characteristics of the designed derivatives. BIOVIA Discovery Studio Visualiser was used to create molecular visualisation and protein-ligand binding orientations that allowed easy visualisation of the ligand-protein interactions and identification of the important binding site residues (Yavuz, 2024).
RESULTS
Drug-Likeness and Bioactivity
Molinspiration analysis revealed that all the designed Mannich bases substituted benzimidazole derivatives complied with Lipinski’s Rule of Five Table 1. The molecular weight was between 281.31 and 436.49 g/mol. The donor and acceptors of hydrogen bonds were acceptable. The TPSA ranged from 67.15 to 127.32Å. The prediction of bioactivity showed that all derivatives had high scores on enzyme inhibition Table 2. The compounds were moderately active to GPCR ligands, ion channel modulators, kinase inhibitors, nuclear receptor ligands, and protease inhibitors.
| Code | miLogP | TPSA | nAtoms | MW (gm) | nON | nOHNH | Nviolation | Nrotb | Vol |
|---|---|---|---|---|---|---|---|---|---|
| 1AA | 2.96 | 67.15 | 21 | 281.31 | 5 | 2 | 0 | 4 | 253.42 |
| 1BA | 4.99 | 67.15 | 26 | 343.39 | 5 | 2 | 0 | 5 | 308.27 |
| 1SA | 4.73 | 87.38 | 27 | 359.38 | 6 | 3 | 0 | 5 | 316.29 |
| 1PA | 4.52 | 104.45 | 29 | 387.39 | 7 | 3 | 0 | 6 | 335.27 |
| 1ASA | 4.5 | 93.46 | 30 | 401.42 | 7 | 2 | 0 | 7 | 352.8 |
| 2AA | 1.75 | 90.02 | 22 | 316.39 | 6 | 3 | 0 | 4 | 269.14 |
| 2BA | 3.78 | 90.02 | 27 | 378.46 | 6 | 3 | 0 | 5 | 323.99 |
| 2SA | 3.51 | 110.25 | 28 | 394.46 | 7 | 4 | 0 | 5 | 332 |
| 2PA | 3.3 | 127.32 | 30 | 422.47 | 8 | 4 | 0 | 6 | 350.99 |
| 2ASA | 3.28 | 116.32 | 31 | 436.49 | 8 | 3 | 0 | 7 | 368.52 |
| Code | Bioactivity score for various targets | |||||
|---|---|---|---|---|---|---|
| GPCR | ICM | KI | NRL | PI | EI | |
| 1AA | -0.2 | -0.22 | -0.25 | -0.46 | -0.52 | -0.01 |
| 1BA | -0.12 | -0.1 | -0.05 | -0.17 | -0.3 | 0.1 |
| 1SA | -0.12 | -0.14 | -0.06 | -0.13 | -0.32 | 0.11 |
| 1PA | -0.04 | -0.09 | -0.04 | -0.11 | -0.24 | 0.11 |
| 1ASA | -0.17 | -0.17 | -0.17 | -0.11 | -0.31 | 0.1 |
| 2AA | -0.29 | -0.31 | -0.18 | -0.84 | -0.31 | 0.07 |
| 2BA | -0.21 | -0.18 | -0.02 | -0.52 | -0.17 | 0.16 |
| 2SA | -0.2 | -0.2 | -0.02 | -0.45 | -0.17 | 0.18 |
| 2PA | -0.11 | -0.16 | -0.05 | -0.31 | -0.09 | 0.21 |
| 2ASA | -0.26 | -0.24 | -0.15 | -0.42 | -0.19 | 0.15 |
Toxicity
OSIRIS Property Explorer analysis indicated that there was no significant risk of mutagenic, tumorigenic, irritant, and reproductive toxicity Table 3.
| Code | C LogP | Solubility | Drug likeness | Drug score | Mutagenic Effect | Tumorigenic Effect | Irritant Effect | Reproductive Effect |
|---|---|---|---|---|---|---|---|---|
| 1AA | 2.53 | -2.66 | 1.29 | 0.79 | ||||
| 1BA | 3.94 | -5.17 | 2.19 | 0.56 | ||||
| 1SA | 3.59 | -4.88 | 1.27 | 0.57 | ||||
| 1PA | 3.43 | -5.19 | -1.4 | 0.36 | ||||
| 1ASA | 3.93 | -5.47 | 1.82 | 0.5 | ||||
| 2AA | 1.98 | -2.54 | 3.93 | 0.88 | ||||
| 2BA | 3.39 | -5.06 | 4.78 | 0.61 | ||||
| 2SA | 3.04 | -4.76 | 3.94 | 0.64 | ||||
| 2ASA | 3.38 | -5.36 | 4.35 | 0.54 | ||||
| Ciprofloxacin (Standard) | -1.53 | -3.32 | 2.07 | 0.82 |
ADMET and Pharmacokinetic Properties
The outcomes of Swiss ADME revealed positive gastrointestinal absorption of most of the compounds except 2PA Table 4. Compounds 1AA and 1BA had the lowest TPSA (67.15 Å) and were predicted to penetrate the Blood Brain Barrier. It was estimated that the rest of the derivatives would possess low BBB permeability. Profiling of CYP inhibition indicated possible CYP2D6 inhibition of some of the derivatives, but lacked expected CYP inhibition of 1PA, 1ASA, and 2ASA. The values of skin permeability (Log Kp) were between -5.31 and -6.82 cm/s. The bioavailability scores were 0.55-0.56.
| Code | Physicochemical Properties | Pharmacokinetics Properties | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| MW | RB | TPSA (Å) | ESOL Class | GIA | BBB P | Pgp S | CYP1 | CYP2 | Log Kp (cm/s) | |
| 1AA | 281.31 | 4 | 67.15 | Sol | High | Yes | No | Yes | Yes | -5.83 |
| 1BA | 343.39 | 5 | 67.15 | M Sol | High | Yes | No | Yes | Yes | -5.31 |
| 1SA | 359.38 | 5 | 87.38 | M Sol | High | No | No | Yes | Yes | -5.67 |
| 1PA | 387.39 | 6 | 104.45 | M Sol | High | No | No | No | No | -5.92 |
| 1ASA | 401.42 | 7 | 93.46 | Sol | High | No | No | No | No | -5.84 |
| 2AA | 316.39 | 4 | 90.02 | M Sol | High | No | No | Yes | Yes | -6.73 |
| 2BA | 378.46 | 5 | 90.02 | M Sol | High | No | No | Yes | Yes | -6.21 |
| 2SA | 394.46 | 5 | 110.25 | M Sol | High | No | No | Yes | Yes | -6.56 |
| 2PA | 422.47 | 6 | 127.32 | M Sol | Low | No | No | Yes | No | -6.82 |
| 2ASA | 436.49 | 7 | 116.32 | M Sol | High | No | No | No | No | -6.74 |
Lipophilicity and Medicinal Chemistry Evaluation
The lipophilicity parameters (iLOGP, XLOGP, WLOGP, MLOGP, consensus LogP) were acceptable drug-like, Table 5. No PAINS notifications were identified. Brenk alerts were minimal. The score of synthetic accessibility was 2.39 to 3.17, Table 6.
| Code | Lipophilicity | Drug Likeness | BAS | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| No. of Violations | |||||||||||
| I Log P | X Log P | W Log P | M Log P | Cons. Log P | L | G | V | E | M | ||
| 1AA | 1.91 | 3.08 | 2.77 | 1.32 | 2.21 | 0 | 0 | 0 | 0 | 0 | 0.85 |
| 1BA | 2.43 | 4.34 | 4.13 | 2.27 | 3.24 | 0 | 0 | 0 | 0 | 0 | 0.85 |
| 1SA | 2.47 | 3.98 | 3.83 | 1.72 | 2.91 | 0 | 0 | 0 | 0 | 0 | 0.56 |
| 1PA | 2 | 3.86 | 3.83 | 1.88 | 2.81 | 0 | 0 | 0 | 0 | 0 | 0.56 |
| 1ASA | 2.76 | 4.09 | 4.05 | 2.1 | 3.2 | 0 | 0 | 0 | 0 | 0 | 0.56 |
| 2AA | 1.61 | 2.11 | 2.8 | 1.29 | 1.69 | 0 | 0 | 0 | 0 | 0 | 0.55 |
| 2BA | 2.2 | 3.38 | 4.16 | 2.27 | 2.74 | 0 | 0 | 0 | 0 | 0 | 0.55 |
| 2SA | 1.66 | 3.02 | 3.86 | 1.74 | 2.29 | 0 | 0 | 0 | 0 | 0 | 0.55 |
| 2PA | 1.42 | 2.9 | 3.86 | 2.18 | 2.3 | 0 | 0 | 0 | 1 | 0 | 0.56 |
| 2ASA | 2.6 | 3.13 | 4.08 | 2.4 | 2.78 | 0 | 0 | 0 | 0 | 0 | 0.55 |
| Code | PAINS #alerts | Brenk #alerts | Leadlikeness #alerts | Synthetic Accessibility |
|---|---|---|---|---|
| 1AA | 0 | 0 | Yes | 2.39 |
| 1BA | 0 | 0 | No | 2.76 |
| 1SA | 0 | 0 | No | 2.8 |
| 1PA | 0 | 0 | No | 2.97 |
| 1ASA | 0 | 1 | No | 3.04 |
| 2AA | 0 | 0 | Yes | 2.68 |
| 2BA | 0 | 0 | No | 2.96 |
| 2SA | 0 | 0 | No | 2.97 |
| 2PA | 0 | 0 | No | 3.08 |
| 2ASA | 0 | 1 | No | 3.17 |
Molecular Docking
The derivatives were successfully docked into the active site of DNA gyrase (PDB ID: 1KZN). The binding energies ranged from -6.0 to -9.24 kcal/mol Table 7. The binding affinities of 2BA and 2ASA were the highest (-9.24 and -9.06 kcal/mol, respectively) and were greater than that of ciprofloxacin (-7.33 kcal/mol). Hydrogen bond interactions were found with Val120, Asn46, Thr165, and Gly119. 3.93 and 3.71 Å were also observed to have a π-sigma interaction. Other π-alkyl contacts with Ile78, Val167, and Ala47 and a π-anion interaction with Gln50 were also found Figures 4-6.
| Code | Binding Energy (Kcal/mol) | Ligand Efficiency | Inhibitory Constant | Vander waals Desolvation Energy |
|---|---|---|---|---|
| 1AA | -6 | -0.29 | 40.09µM | -7.09 |
| 1BA | -7.44 | -0.29 | 3.53µM | -9.24 |
| 1SA | -7.4 | -0.27 | 3.79µM | -9.02 |
| 1PA | -6.76 | -0.23 | 11.16µM | -9.02 |
| 1ASA | -6.94 | -0.23 | 8.13µM | -9.26 |
| 2AA | -7.22 | -0.33 | 5.08µM | -8.13 |
| 2BA | -9.24 | -0.34 | 169.77nM | -10.78 |
| 2SA | -8.75 | -0.31 | 384.59nM | -10.25 |
| 2PA | -8.2 | -0.27 | 981.53nM | -10.1 |
| 2ASA | -9.06 | -0.29 | 228.36nM | -11.21 |
| Ciprofloxacin (Standard) | -7.33 | -0.31 | 4.27µM | -7.76 |
DISCUSSION
The compliance with Lipinski’s criteria indicates that the designed derivatives possess favorable oral drug-likeness characteristics. The observed TPSA range suggests an optimal balance between polarity and membrane permeability, which is essential for intestinal absorption. The strong enzyme inhibition scores support the rational design strategy targeting DNA gyrase. Minimal toxicity risks and acceptable drug scores suggest a favorable safety profile. High predicted gastrointestinal absorption and moderate bioavailability scores further support oral administration potential. Limited BBB permeability for most compounds may reduce central nervous system-related adverse effects. The absence of predicted CYP inhibition in selected derivatives (1PA, 1ASA, 2ASA) suggests reduced risk of drug-drug interactions.
DNA gyrase was selected as the primary molecular target in the present study because it plays a crucial role in bacterial DNA replication, transcription, and chromosome segregation, making it an essential enzyme for bacterial survival. Inhibition of DNA gyrase has been widely exploited in antibacterial therapy, as demonstrated by the success of fluoroquinolone antibiotics. Moreover, several studies have reported that benzimidazole derivatives exhibit inhibitory activity against bacterial DNA gyrase, which provided a strong rationale for selecting this enzyme for molecular docking analysis. Although other microbial targets such as topoisomerase and Dihydrofolate Reductase (DHFR) are also important in antimicrobial drug discovery, the current study focused on DNA gyrase to specifically evaluate the interaction potential of the designed Mannich base-substituted benzimidazole derivatives with this validated target. Future studies may include multi-target docking and experimental validation against additional enzymes such as topoisomerase and DHFR to further broaden the antimicrobial evaluation of these compounds.
Substituents at the R position play a crucial role in modulating activity. Derivatives containing aromatic or electron-withdrawing substituents, such as Carboxyphenyl (C₆H₄COOH), Hydroxy Phenyl (C₆H₄OH), and Acetoxyphenyl (CH₃COOC₆H₄) groups, exhibited improved docking scores compared with simple electron releasing substituents like alkyl/ aryl substituents [methyl (CH₃) and Phenyl (C6H5)]. These aromatic substituents increase hydrophobic interactions and π-stacking interactions within the enzyme binding pocket. In particular, the presence of polar functional groups such as -COOH and -OH enhances hydrogen bonding with key residues including Val120, Asn46, Thr165, and Gly119, thereby stabilizing the ligand-protein complex.
The R₁ substituent also significantly affects the binding interactions. Compounds containing Sulfonamide (C₆H₄SO₂NH₂) groups showed stronger binding affinity due to their ability to act as hydrogen bond donors and acceptors. This facilitates stronger hydrogen bonding with amino acid residues in the active site and contributes to improved ligand stabilization.
Among the designed compounds, 2BA and 2ASA showed the highest binding affinities (-9.24 and -9.06 kcal/mol, respectively). The enhanced activity of these compounds can be attributed to the combined effect of aromatic substituents, hydrogen bonding capability, and favorable hydrophobic interactions. Docking analysis indicated that these derivatives form multiple stabilizing interactions including hydrogen bonds, π-sigma interactions, π-alkyl contacts with residues such as Ile78, Val167, and Ala47, and π-anion interactions with Gln50.
Overall, the SAR analysis suggests that increasing aromaticity and introducing polar functional groups capable of hydrogen bonding significantly enhance the inhibitory potential of benzimidazole-Mannich derivatives against DNA gyrase. These findings highlight compounds 2BA and 2ASA as promising lead molecules for further optimization and antimicrobial development.
CONCLUSION
In this study, a series of Mannich base-substituted benzimidazole derivatives were designed and evaluated using computational approaches to explore their potential as antimicrobial agents targeting DNA gyrase. Drug-likeness analysis indicated that all the designed compounds complied with Lipinski’s Rule of Five, suggesting favorable physicochemical properties and good oral bioavailability. Bioactivity prediction further supported their potential as enzyme inhibitors.
Toxicity assessment using OSIRIS Property Explorer revealed minimal risks of mutagenicity, tumorigenicity, irritancy, and reproductive toxicity for most of the derivatives. In addition, SwissADME analysis demonstrated acceptable pharmacokinetic properties, including high gastrointestinal absorption for most compounds and suitable lipophilicity values. These findings suggest that the designed molecules possess promising drug-like characteristics.
Molecular docking studies showed that all the derivatives interacted effectively with the active site of DNA gyrase (PDB ID: 1KZN). The binding energies ranged from -6.0 to -9.24 kcal/mol, indicating favorable ligand-protein interactions. Among the compounds, 2BA and 2ASA exhibited the strongest binding affinities, which were higher than that of the reference drug ciprofloxacin. The enhanced activity of these compounds can be attributed to the presence of aromatic substituents and functional groups capable of forming hydrogen bonds and hydrophobic interactions with key amino acid residues within the enzyme binding pocket.
Overall, the results suggest that benzimidazole-Mannich base derivatives represent promising scaffolds for the development of new antimicrobial agents targeting DNA gyrase. Among the designed compounds, 2ASA and 2BA emerged as potential lead molecules with favorable binding affinity and pharmacokinetic properties. However, further in vitro and in vivo studies are necessary to validate their antimicrobial activity and therapeutic potential.