Molecular Docking Study of Benzimidazoles against ?-Catenin: In Silico Approach to Anticancer Drug Discovery

Molecular docking is a computational technique used to predict the most favourable orientation of one molecule relative to another during the formation of a stable complex with minimum overall energy. Although this interaction is a natural process within cells, in computational terms, molecular docking refers to the analysis of how two or more molecular structures fit together. It involves using molecular modelling software to position a molecule within the binding site(s) of a target in order to determine the optimal interaction and the resulting complex structure. This approach examines the possible conformations a ligand may adopt within the binding pocket(s) of a macromolecular target and estimates the binding free energy of the ligand-receptor complex. This estimation considers key phenomena involved in intermolecular recognition. Molecular docking operates on the principle of molecular complementarity, widely used in drug discovery, molecular docking has been a key tool since the early 1980s. Today, the field of structural biology is advancing rapidly and supports sophisticated computational techniques such as structure-based virtual high-throughput screening (vHTS). These methods are employed in hit identification and lead optimization during the drug discovery and development process ^[1].

Cancer is classified as a hyperproliferative disorder, encompassing a range of conditions characterized by the abnormal transformation of cells, which leads to uncontrolled proliferation, tissue invasion, and systemic dissemination through the lymphatic system or bloodstream. A key contributing factor in many cancers is the failure of cells to undergo apoptosis, a vital cellular mechanism responsible for regulating cell growth and maintaining tissue homeostasis. The current arsenal of cancer treatments includes surgical resection, radiotherapy, chemotherapy, immunotherapy which stimulates the immune system to eliminate tumour cells and provide long-term protection and molecular targeted therapies, such as tyrosine kinase inhibitors, interferons, and lymphokines, which act specifically on cellular pathways involved in tumour growth and maintenance. Globally, cancer ranks as the second leading cause of death after cardiovascular disease, accounting for approximately 9.6 million deaths annually. Despite significant advancements in the development of anticancer therapeutics, the field continues to face substantial challenges. These include the emergence of drug resistance, limited therapeutic efficacy, and high toxicity levels, all of which significantly affect the quality of life for cancer patients. Consequently, the discovery and development of highly selective, effective, and minimally toxic anticancer agents remains a critical priority in modern cancer research ^[2].

The Wnt signalling system represents a complex group of signal transduction pathways, broadly categorized into two types: canonical (β-catenin-dependent) and non-canonical (β-catenin-independent) pathways. The canonical Wnt pathway, also known as the Wnt/β-catenin signalling pathway, plays a pivotal role in controlling various cellular functions such as proliferation, differentiation, and gene expression. This pathway is primarily dependent on the regulation of β-catenin, a key intracellular signalling molecule. Wnt proteins are secreted glycoproteins that function as the initial signalling molecules in this pathway. Their biological activity depends on lipid modification specifically, acylation which occurs in the endoplasmic reticulum via the enzyme PORCN (Porcupine O-acyltransferase). This lipid modification is crucial for the proper interaction of Wnt proteins with their surface receptors. The non-canonical Wnt pathways primarily consist of two branches: the Wnt/Planar Cell Polarity (PCP) pathway and the Wnt/Ca²? signalling pathway. The Wnt/Ca²? pathway influences cellular behaviour by regulating gene expression associated with cell adhesion, which it does through the induction of intracellular calcium (Ca²?) release. Activation of these non-canonical signalling cascades generally occurs through non-canonical Wnt ligands, such as Wnt5a and Wnt11, which bind to Frizzled (Fzd) receptors. This interaction initiates downstream signalling events that govern various cellular responses distinct from those regulated by the canonical β-catenin pathway ^[3].

Fig 01: Wnt signalling pathway

Colorectal cancer (CRC) is a prevalent malignancy of the digestive system and has shown a rising incidence trend globally. It remains a major contributor to cancer-related mortality, ranking fourth in occurrence and second in deaths worldwide. According to global cancer statistics from 2020, CRC accounted for approximately 10% of all new cancer cases and 9.4% of cancer-related deaths. Standard treatments, including surgical resection, chemotherapy, and radiotherapy, have improved patient outcomes; however, the risk of recurrence and distant metastasis continues to pose significant clinical challenges. A substantial body of research has identified the Wnt/β-catenin signalling pathway as a critical player in CRC pathogenesis. Consequently, targeting the Wnt/β-catenin pathway has emerged as a promising therapeutic strategy. Recent advances in molecular oncology have led to the development of several inhibitors aimed at modulating this pathway, with the goal of suppressing tumour growth and metastasis. As research continues, the Wnt/β-catenin axis remains a significant focus for novel drug discovery and targeted cancer therapies in CRC management ^[4].

Fig 02: Colorectal cancer

Benzimidazole is a bicyclic heteroaromatic compound formed by the fusion of a benzene ring with an imidazole ring. It is a key structural component of vitamin B12 and contains two nitrogen atoms within its ring system, making it a heterocyclic aromatic compound of significant biological relevance. Benzimidazole serves as a versatile pharmacophore and has demonstrated a wide range of therapeutic activities, including anticancer, antimicrobial, anti-inflammatory, antiviral, antihypertensive, antihistaminic, antitubercular, antiulcer, analgesic, anthelmintic, antiprotozoal, antiamoebic, anticonvulsant, and antiparasitic effects. Numerous clinically important drugs incorporate the benzimidazole scaffold due to its favourable biological activity. Particularly noteworthy are its applications in antifungal agents like benomyl and anticancer agents such as nocodazole (a mitotic inhibitor) and veliparib (a PARP inhibitor). Compounds like carbendazim and dovitinib, which also contain benzimidazole moieties, have shown potential against various cancer cell lines, further underscoring the pharmacological importance of this structural moiety ^[5].

Various substituted benzimidazoles have been investigated in preclinical studies for their potential to inhibit FtsZ activity. Among these, benzo[d]imidazole-2-carboxamide derivatives have emerged as particularly promising candidates ^[6]. Numerous studies have demonstrated that benzimidazole derivatives exert anticancer effects through multiple mechanisms. These include DNA intercalation, inhibition of topoisomerases I and II, antagonism of androgen receptors, suppression of enzymes such as poly (ADP-ribose) polymerase (PARP) and dihydrofolate reductase (DHFR), as well as disruption of microtubule function. Additionally, several benzimidazole-based compounds have been identified as effective inhibitors of the epidermal growth factor receptor ^[7].

METHODOLOGY

Twenty benzimidazole derivatives were selected for computational evaluation. Their chemical structures were initially sketched and geometrically optimized using ChemDraw Pro 8.0 to obtain precise conformations suitable for further analysis. Molecular docking was performed employing Auto Dock Vina through the PyRx Virtual Screening Tool, enabling the prediction of binding affinities between the designed compounds and the selected target protein.

Preparation of Ligands

To prepare a ligand using Discovery Studio, begin by launching the software and drawing the ligand using the chemdraw pro v8.0 software and then importing it in formats like. mol, .sdf, .pdb, or. mol2 in discovery studio. Once the ligand structure is available, convert it into a 3D model (if sketched) and use the “Clean Geometry” function to optimize its 3D conformation. Next, add missing hydrogen atoms via the “Add Hydrogens” tool to complete the structure. Then, use the “Prepare Ligands” protocol found under the Small Molecules section, where you can generate tautomer’s, ionization states, and stereoisomers if required. Finally, save the prepared ligand in a compatible format like. mol2 or .pdb for further studies such as molecular docking.

Table 1: Novel Benzimidazole compounds

Compound Code	R1	R2	R3
MN01	H	Cl	H
MN02		Cl	H
MN03	CF3	Cl	H
MN04	H		H
MN05	H
MN06	H
MN07	H
MN08	H
MN09	H		H
MN10	H		H
MN11	H		H
MN12	H		H
MN13			H
MN14			H
MN15			H
MN16			H
MN17			H
MN18	CF3		H
MN19	CF3		H
MN20	CF3		H

The protein used in the study was identified using the NCBI (National Centre for Biotechnology Information) database ^[8]. A search was conducted for the target protein involved in the β-catenin signalling pathway, with Homo sapiens selected as the organism. The amino acid sequence was retrieved in FASTA format. To identify the corresponding Protein Data Bank (PDB) ID, targeting the standard PDB protein database. The resulting hits were filtered based on an E-value of zero and a query coverage between 85% and 100%. The shortlisted PDB IDs were then examined in the RCSB PDB (Research Collaboratory for Structural Bioinformatics) database, with attention to structural quality parameters such as X-ray diffraction resolution, Ramachandran plot validation, presence of mutations, chain selection, and any bound interacting ligands ^[9]. After evaluating these criteria, PDB ID: 8Z10 was chosen for its suitability and downloaded in .pdb format for subsequent molecular docking analysis. Then, the 8Z10 protein structure was imported into Discovery Studio, where water molecules, heteroatoms, existing ligands, and any non-essential chains were removed. Polar hydrogen atoms were then added to the refined protein structure to prepare it for docking studies. The finalized structure was saved in .pdb format for further analysis.

Fig 04: Selection of target protein (PDB ID: 8Z10)

Fig 05: Purified target protein (PDB ID: 8Z10)

Docking Procedure

Virtual screening was carried out using PyRx software, which integrates both Auto Dock 4.2 and Auto Dock Vina for molecular docking simulations. The benzimidazole derivatives were docked with the β-catenin pathway protein (PDB ID: 8Z10). PyRx also includes Open Babel, which facilitates file format conversions. Prior to docking, the protein was prepared and analysed using Discovery Studio. The finalized protein structure in .pdb format was imported into PyRx and converted to. pdbqt format using the "Make Molecule" function. Subsequently, all ligand structures were added and processed one by one. Each ligand underwent energy minimization and was saved in. pdbqt format. Once both protein and ligands were ready, the active site of the protein was defined, and a grid box was positioned to encompass the predicted binding pocket. The grid box coordinates were set at X = 12, Y = -25, and Z = 44. The docking results were saved in a CSV (Comma-Separated Values) file for further evaluation of binding poses and binding affinities. Finally, the docked complexes were visualized and analysed using Biovia Discovery Studio to interpret the interactions.

Drug likeliness

SwissADME software was utilized to evaluate the drug-likeness of the compounds based on Lipinski’s Rule of Five. For this analysis, key 2D molecular descriptors were considered, including molecular weight (MW), number of hydrogen bond acceptors (HBA), number of hydrogen bond donors (HBD), rotatable bonds (RB), lipophilicity (Log P), and molecular surface area. These parameters helped in assessing the suitability of the compounds as potential drug candidates ^[10].

RESULTS AND DISCUSSION

Docking score analysis

Molecular docking was conducted using the β-catenin target protein with PDB ID: 8Z10 to evaluate the binding affinities of a series of benzimidazole derivatives, labelled MN01 through MN20. The results were compared with the standard drug, Teniposide. Among all compounds, MN17 exhibited the most favourable binding energy at –8.6 kcal/mol, followed closely by MN16 (–8.4 kcal/mol), MN06 and MN08 (both –8.1 kcal/mol), and MN05 (–8.0 kcal/mol), matching the docking score of Teniposide. Additional compounds showing strong interactions included MN20 (–7.7 kcal/mol), MN18 (–7.4 kcal/mol), MN14 and MN07 (–7.3 kcal/mol), and MN13 (–7.2 kcal/mol). Moderate binding affinities were observed for compounds such as MN04 (–6.9 kcal/mol), MN10, MN11, and MN15 (each –6.7 kcal/mol), MN19 (–6.6 kcal/mol), and MN12 (–6.3 kcal/mol). Compounds MN02 (–6.0 kcal/mol), MN09 (–5.9 kcal/mol), and MN03 (–5.7 kcal/mol) demonstrated comparatively weaker interactions, with MN01 showing the lowest binding energy at –4.6 kcal/mol. These findings suggest that several of the designed compounds, particularly MN17 and MN16, may possess superior binding affinity compared to the reference compound Teniposide (-8.0 kcal/mol) and could be considered promising candidates for further investigation.

Table 2: Docking Score and Pharmacokinetic properties of novel compounds

A comprehensive evaluation of the ADME (Absorption, Distribution, Metabolism, and Excretion) profiles of 20 novel compounds (MN01–MN20) along with Teniposide reveals distinct pharmacokinetic behaviours relevant to drug development. Among the studied compounds, 11 demonstrated high gastrointestinal (GI) absorption, indicating favourable oral bioavailability, while the remaining 9, including Teniposide, exhibited low GI absorption. Blood-brain barrier (BBB) permeability was observed in 7 compounds (MN01–MN03, MN09–MN12), suggesting potential central nervous system (CNS) activity, whereas Teniposide and the other molecules lacked BBB permeation. Regarding efflux susceptibility, 7 compounds (MN02, MN09–MN12, MN15, and MN20) were identified as P-glycoprotein (P-gp) substrates, potentially limiting their intracellular concentrations due to active efflux mechanisms. The metabolic liability of these compounds was assessed through cytochrome P450 (CYP) enzyme inhibition. Notably, CYP1A2 inhibition was the most prevalent, observed in 17 out of 20 compounds, which may raise concerns about drug–drug interactions. CYP2C19 was inhibited by 11 compounds, while CYP3A4 inhibition was reported in 5 compounds, reflecting moderate metabolic involvement. In contrast, CYP2C9 and CYP2D6 were less commonly inhibited, with Teniposide and MN08 affecting CYP2C9, and only MN12 inhibiting CYP2D6. Overall, the dataset indicates that while many compounds exhibit promising absorption and CNS permeability, most also inhibit key CYP enzymes, especially CYP1A2, which could pose metabolic interaction risks. Teniposide, despite showing low absorption and no BBB permeability or P-gp substrate activity, displays a cleaner CYP inhibition profile, targeting only CYP2C9, thus it is not orally active and administered through intravenously. These findings support the need for cautious optimization of ADME properties in early drug development to balance efficacy with pharmacokinetic safety.

Table 3: Lipinski’s RO5