Skip to main content
NCBI home page
In Silico Pharmacology logoLink to In Silico Pharmacology
. 2024 Oct 29;12(2):95. doi: 10.1007/s40203-024-00276-3

Glucosinolates and Indole-3-carbinol from Brassica oleracea L. as inhibitors of E. coli CdtB: insights from molecular docking, dynamics, DFT and in vitro assay

Faria Tasnim 1,#, Md Eram Hosen 1,✉,#, Md Enamul Haque 2, Ariful Islam 3, Mst Naharina Nuryay 4, Jannatul Mawya 5, Najnin Akter 5, Delara Yesmin 4, Md Mosabbir Hossain 4, Nilima Rahman 5, B M Mahmudul Hasan 6, Md Naimul Hassan 7, Md Mahmudul Islam 1, Md Khalekuzzaman 3
PMCID: PMC11519271  PMID: 39479380

Abstract

Escherichia coli (E. coli), a common human gut bacterium, is generally harmless but capable of causing infections and contributing to diseases like urinary tract infections, sepsis/meningitis, or diarrheal diseases. Notably, E. coli is implicated in developing gallbladder cancer (GBC) either through ascending infection from the gastrointestinal tract or via hematogenous spread. Certain E. coli strains are known to produce toxins, such as cytolethal distending toxins (CDTs), that directly contribute to the genetic mutations and cellular abnormalities observed in GBC. Broccoli (Brassica oleracea) is known for its health-promoting properties, including antimicrobial, antioxidant, and immunomodulatory effects, and is rich in essential compounds. Our study investigates the potential of the phytochemicals of B. oleracea to inhibit the CdtB (PDB ID: 2F1N) protein of E. coli which plays a significant role in the pathogenesis of GBC. By employing in silico molecular docking, Glucosinolates and Indole-3-carbinol emerged as promising inhibitors, demonstrating strong bonding affinities of -8.95 and − 8.5 Kcal/mol, respectively. The molecular dynamic simulation showed that both compounds maintained stable interaction with CdtB with minimal conformational changes observed in the protein-ligand complexes. Additionally, the ADMET analysis provided evidence for the drug-likeness properties of the lead compounds. Furthermore, the DFT (Density Functional Theory) revealed that Indole-3-carbinol is more chemically stable but less reactive than Glucosinolates, with HOMO-LUMO gaps of 5.14 eV and 4.50 eV, respectively. Finally, the in vitro antibacterial assessment confirmed the inhibitory effect of Glucosinolates and Indole-3-carbinol against E. coli through disc diffusion assay with the zone of inhibition 34.25 ± 0.541 and 28.67 ± 0.376 mm compared to the control ciprofloxacin. Our study provides crucial data for developing novel therapeutic agents targeting E. coli-associated GBC from the phytochemicals of B. oleracea.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40203-024-00276-3.

Keywords: E. coli, Cytolethal distending toxins (CDTs), Gallbladder cancer, Brassica oleracea, Molecular docking, Molecular dynamics, DFT, In vitro antibacterial activity

Introduction

Escherichia coli (E. coli) is a commonly found predominant facultative anaerobe of human and animal intestinal flora (Peng et al. 2024). This bacterium typically colonizes the gastrointestinal tract (GIT) of infants within hours of birth, forming a beneficial relationship with the host (Martinson and Walk 2020). E. coli is usually harmless in the intestinal lumen (Nougayrède et al. 2021) but can cause infections in immunocompromised individuals or those with compromised GIT barriers (Pokharel et al. 2023). However, healthy individuals can also be infected by highly adapted E. coli strains, which have evolved to cause various diseases, including urinary tract infections, sepsis/meningitis, and diarrheal diseases, depending on their location or spread within the body (Bendary et al. 2022). E. coli has been increasingly recognized for its potential role in developing cancers, including gallbladder cancer (GBC) (Eyvazi et al. 2020). The gallbladder can become infected by E. coli through an ascending infection from the GIT or hematogenous spread (Hynes et al. 2020). In the ascending pathway, E. coli migrates from the intestines through bile ducts into the gallbladder, especially when disruptions in biliary anatomy or conditions like gallstones facilitate colonization (Wang et al. 2024). Alternatively, E. coli may enter the bloodstream from an initial infection site (intestines, urinary tract) and reach the gallbladder via the hepatic artery (Tsuchiya et al. 2018). Once inside the biliary system, E. coli can adhere to and invade the gallbladder’s epithelial cells, causing cholecystitis and potentially leading to chronic infection (Tsuchiya et al. 2018). Such a chronic inflammatory environment creates conditions conducive to the development of cancerous cells. Additionally, certain E. coli strains produce toxins, such as cytolethal distending toxins (CDTs), that directly contribute to the genetic mutations and cellular abnormalities observed in GBC (Eyvazi et al. 2020). CDT was the first identified bacterial genotoxin encoding three polypeptides: CdtA, CdtB, and CdtC. This toxin causes double-strand breaks (DSBs) in DNA, activating the ataxia telangiectasia mutated (ATM)-dependent DNA damage response, which results in the formation of DNA repair complexes, ultimately causing irreversible cell cycle arrest at the G2/M phase and inducing apoptosis (Liu et al. 2024). Among the components, CdtB is the active subunit, functionally similar to mammalian deoxyribonuclease I (DNase I) (Pons et al. 2021).

Ciprofloxacin, cefepime, and cefazolin are commonly used antibiotics for treating E. coli infections (Eissa 2024). However, E. coli has become more resistant to various antibiotics, including fluoroquinolones, beta-lactams, and aminoglycosides, due to selection pressure from the extensive use of antibiotics (Foudraine et al. 2021). Multidrug-resistant (MDR) E. coli strains, including those producing extended-spectrum beta-lactamase (ESBL), pose significant challenges by reducing available treatment options and increasing morbidity and mortality (Bitew and Tsige 2020). Despite the efficacy of synthetic medications, they often have serious side effects. Therefore, it is imperative to explore alternative treatments that minimize side effects while effectively combating E. coli. This has prompted researchers to consider plant-based treatments for E. coli infection due to their minimal side effects (Cock et al. 2021).

Broccoli, a green plant from the cabbage family (Brassicaceae), is categorized within the Italica cultivar group of the Brassica oleracea species (Li et al. 2022). This vegetable is recognized for its pharmaceutical significance due to various health-promoting properties, including antimicrobial, antioxidant, anticancer, immunomodulatory, antidiabetic, hepatoprotective, cardioprotective, and anti-amnesic effects (Syed et al. 2023). Broccoli is particularly rich in essential compounds like polyphenols, glucosinolates, sulforaphane, and selenium (Bianchi et al. 2024).

The advancement of medical research has led to innovative antibiotic drugs with potential anticancer properties (Bateman and Conrads 2018). These therapies aim to combat bacterial infections while displaying anticancer effects. Our current investigation focuses on identifying inhibitors targeting the CdtB (2F1N) proteins of E. coli from phytochemicals in broccoli. The primary objective is to identify promising lead compounds capable of inhibiting E. coli pathogenesis, potentially serving as a therapeutic opportunity for E. coli-associated GBC in humans.

Materials and methods

The overall working flowchart of the current study is shown in Fig. 1.

Fig. 1.

Fig. 1

Open in a new tab

The overall working flow diagram of the present study

Collection and preparation of Brassica oleracea phytochemicals

A total of 58 chemical compounds of Brassica oleracea were retrieved in SDF format from the Pub-Chem database (https://pubchem.ncbi.nlm.nih.gov/). We meticulously refined the data for accuracy and subsequent analysis, furthermore used Avogadro software (version 1.2.0) to optimize their energy via the mmf94 force field) (Hanwell et al. 2012).

Protein preparation

The x-ray crystallography structure of CdtB (PDB: 2F1N) proteins of E. coli was retrieved from the protein data bank (Burley et al. 2017). We initially used Discovery Studio software (version 21.1.0.0) to clean up the protein structures and remove any irrelevant molecules (Inc 2012). Then, energy minimization and optimization of the cleaned proteins were done using the GROMOS96 43b1 force field, aided by the SwissPDB Viewer software (version 4.1) (Guex and Peitsch 1997).

Molecular docking

The molecular docking was performed between the plant chemicals found in B. oleracea and the target proteins (CdtB, PDB ID: 2F1N) from E. coli bacteria. For this, we used a software program called PyRx (version 0.8) (https://sourceforge.net/projects/pyrx/) that relies on a method called AutoDock Vina to simulate the docking process (Trott and Olson 2010). The molecular docking was performed based on the previous methods with small modifications (Hosen et al. 2023; Kandsi et al. 2022; Kumar Deb et al. 2022). First, the data was prepared by converting the protein structures into a suitable format (macromolecule) and then the plant chemicals (ligands) into a different format (PDBQT). The final docking calculation was conducted and top molecules were selected based on lower binding energy. We then used Discovery Studio software to analyze the specific interactions and how these plant chemicals position themselves when bound to the protein (Inc 2012).

Molecular dynamics

The molecular dynamics simulations were performed using YASARA dynamics software version 19.12.4 and AMBER14 force field (Land and Humble 2018; Salomon-Ferrer et al. 2013). The hydrogen bond network was initially cleaned and optimized alongside the docked complexes. Reduction of protein complexes was achieved using a TIP3P water solvation model (density 0.997 g/L, temperature 25 °C, pressure 1 atm) through the steepest gradient approaches (Harrach and Drossel 2014). Physiological conditions were maintained at 0.9% NaCl, 310 K, and pH 7.4 to neutralize the simulated system (Krieger et al. 2012). The simulation time step was set at 1.25 frames per second. Long-range electrostatic interactions were computed using the particle mesh Ewald (PME) method with an 8.0 Å cutoff radius (Essmann et al. 1995). Trajectories were saved every 100 ps, and the final simulation duration was 100 ns (Krieger and Vriend 2015). The MD.xtc files of all conducted compounds are shown in Supplementary file 1. Analysis of the simulation trajectories included assessing root-mean-square deviation, solvent-accessible surface area, radius of gyration, hydrogen bonding, and root mean square fluctuation (Baildya et al. 2021; Islam et al. 2020).

Binding-free energy calculation using MM/PBSA

Evaluating the strength of the interaction between a drug and a protein is essential, and one crucial method for this assessment is the calculation of binding free energy. This analysis offers valuable insights into the energetic aspects and stability of the drug-protein complex. In YASARA software, the MM-Poisson–Boltzmann surface area (MM-PBSA) method was employed to determine the binding free energy by analyzing various snapshots of the complex. The calculation involved the following formula:

graphic file with name M1.gif

The computations utilized the AMBER14 force field, and YASARA macros were employed to efficiently calculate the binding energy using the MM-PBSA method.

ADMET analysis

Following the successful completion of docking and dynamics studies, phytochemicals were further evaluated through absorption, distribution, metabolism, excretion, and toxicity (ADMET) analysis to ascertain their potential as lead molecules. For this purpose, the Pkcsm and the SwissADME web servers were employed to assess ADMET profiles and determine the adherence of molecules to Lipinski’s rule of five, respectively (Daina et al. 2017; Pires et al. 2015; Pollastri 2010).

In vitro antibacterial activity

Chemical and reagents

Glucosinolates and Indole-3-carbinol compounds were purchased from Sigma-Aldrich as high-performance liquid chromatography (HPLC) standards. HPLC-grade methanol was utilized for sample preparation. Ciprofloxacin, an antibiotic drug, was acquired from Square Pharmaceuticals Ltd. Luria Bertani (LB) broth and LB agar media were obtained from Sigma-Aldrich (USA).

Collection of bacterial samples

Bacteria associated with GBC, specifically E. coli ATCC25922, were obtained from the State Key Laboratory of Microbiology and Bioinformatics, Department of Microbiology, Shaheed Shamsuzzoha Institute of Biosciences, Affiliated with the University of Rajshahi, Rajshahi, Bangladesh. These bacteria were initially clinically isolated from a GBC patient. Upon collection, the bacteria were cultured on LB agar media and allowed to grow overnight at 37 °C. For long-term storage, E. coli was maintained at -80 °C. E. coli is categorized as a Biosafety Level 2 (BSL-2) pathogen, and thus, we strictly adhered to all relevant guidelines and regulations for the proper use and handling of the bacterium.

Determination of in vitro antibacterial activity

The two most promising compounds, Glucosinolates and Indole-3-carbinol, which demonstrated significant inhibitory effects against E. coli ATCC25922 through computational analysis, were selected for further investigation regarding in vitro antibacterial activity against this GBC-associated bacterium. For the preparation of test solutions, both compounds were dissolved in 60% methanol. Following this, an in vitro antibacterial assessment was carried out using the disc diffusion method with slight modifications (Tiruneh et al. 2022), employing the concentrations of 25, 50, and 75 µg/disc. Bacterial cultures were initially grown overnight in nutrient broth at 37 °C with agitation at 180 rpm. Subsequently, the bacterial suspension with a concentration of 1 × 106 CFU/mL was evenly spread onto LB agar plates. For the experiment, Whatman No.1 filter paper discs, each with a diameter of five mm, were used. The discs were impregnated with 25, 50, 75 µg/disc of each compound (Glucosinolates and Indole-3-carbinol) and strategically placed on the agar plates. Ciprofloxacin, an antibiotic drug, served as a positive control. Clear zones around the discs, indicating inhibition of bacterial growth, were assessed after 24 h of incubation. The diameters of these inhibition zones were measured using a millimeter (mm) scale. To ensure precision, the experiment was repeated three times, and the data were subsequently presented as the mean and standard deviation of the results.

DFT study

The electronic properties of the ligands were investigated by analyzing the 3D structures of the phytochemicals using density functional theory. The drug compounds underwent complete optimization utilizing the B3LYP functional and DEF2-SVP basis set (Emad E. El-Katori & Ashraf S. Abousalem, 2019). This approach was employed to gather data on intramolecular charge transfer (ICT) through the examination of the frontier molecular orbitals (FMOs), highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) energies.

Results and discussion

Molecular docking study

Molecular docking, a cornerstone of computational structure-based drug design, elucidates the optimal configuration and binding mode of a small molecule within a protein, streamlining future lead optimization efforts and enhancing the rational design process to fine-tune protein-ligand interactions for improved efficacy and minimized clashes (de Ruyck et al. 2016; Ferreira et al. 2015). E. coli has been implicated in GBC by secretion of virulence factors regulated by Cdt proteins (Nath 2010). Despite the association, there has been a lack of published studies investigating the inhibitory activity of B. oleracea phytochemicals on Cdt proteins. In this study, we conducted an in silico docking investigation targeting CdtB protein (PDB ID: 2F1N) of E. coli, responsible for the expression of virulence genes, using phytochemicals derived from B. oleracea. Among 58 compounds of B. oleracea screened, the top two compounds, Glucosinolates and Indole-3-carbinol, were selected based on the binding energies.

Glucosinolates and Indole-3-carbinol exhibited superior docking scores at -8.95 and − 8.5 kcal/mol, respectively, compared to the control complex’s binding energy of -7.5 kcal/mol (Table 1). The Glucosinolates complex with 2F1N protein also showed supremacy in forming hydrogen bonds at four amino acid residues of ILE239 (1.83 Å), ILE239 (2.64 Å), TYR232 (3.04Å) and ASP242 (3.67Å) (Fig. 2). We also observed some other hydrophobic interactions at different residues, which are critical in stabilizing the complex and influencing binding affinity. The 2F1N + Indole-3-carbinol complex revealed three hydrogen bond interactions with the backbone amino acid residues of TRP10 (3.00Å), THR237 (2.59Å), and ILE239 (5.03Å). Meanwhile, a single pi-sulfur and pi-anion bond was on THR237 and ILE239, respectively. Meanwhile, the control complex showed only one hydrogen bond at THR237 (2.90Å) and four other different bonds with the target protein (Fig. 2).

Table 1.

Interaction of the ligand molecule glucosinolates and Indole-3-carbinol against 2F1N protein of E. Coli mentioning binding energy, non-covalent interaction, interacting amino acids, bond types, and their distance

Complex Binding energy (kcal/mole) Amino acid residues Bond types Distance (Å)
2F1N + glucosinolate -8.95

A: ILE239

A: ILE239

A: TYR232

A: ASP242

A: PHE244

A: PHE244

A: PHE244

H

H

H

H

P-Su

P-P

P-P

1.83

2.64

3.04

3.67

5.35

3.93

3.95
2F1N + indole-3-carbinol -8.5

A: TRP10

A: THR237

A: THR237

A: ILE239

A: SER241

H

H

P-S

H

UDD

3.00

2.59

3.51

5.03

2.01
2F1N + control -7.5

A: THR237

A: PHE244

A: PHE244

A: PHE244

A: PRO245

H

A

A

P-A

P-A

2.90

4.13

5.21

5.41

5.47
Open in a new tab

H: Hydrogen bond; P-P: Pi-pi stacked bond; P-S: Pi-Sigma bond; P-A: Pi-Alkyl; Pi-An: Pi-Anion; P-Su: Pi-Sulfur bond; A: Alkyl bond; UDD: Unfavorable donor-donor

Fig. 2.

Fig. 2

Open in a new tab

Molecular docking interactions of the compound Glucosinolates and Indole-3-carbinol from Brassica oleracea with 2F1N protein of E. coli; pose, and 2D view of compounds

The protein 2F1N has still not been explored as inhibited by phytocompounds through molecular docking studies. Nevertheless, lower binding energy and rigid bonding patterns indicate both Glucosinolates and Indole-3-carbinol efficiently bind with the active site of the 2F1N protein and can cause inhibition.

Molecular dynamic simulation

Molecular Dynamics (MD) simulations were employed to investigate the dynamic behaviors of protein-ligand complexes at the atomic scale, yielding information crucial for drug discovery, elucidating biological processes, and understanding structure-function relationships (Durrant and McCammon 2011; Hollingsworth and Dror 2018; Sargsyan et al. 2017). MD simulations were subsequently conducted on the most promising complexes to validate their stability and rigidity over a 100 nanoseconds (ns) period, leading to the identification of potent inhibitors. Analysis metrics, including Root Mean Square Deviation (RMSD), Solvent Accessible Surface Area (SASA), Radius of Gyration (Rg), Hydrogen Bonding and Root-mean-square-fluctuation (RMSF) events, were utilized to evaluate the complexes, with a focus on the CdtB (PDB ID: 2F1N) protein complex, as illustrated in Figs. 3, and 4.

Fig. 3.

Fig. 3

Open in a new tab

The molecular dynamics simulation study concerning complexes of Glucosinolates and Indole-3-carbinol from Brassica oleracea with 2F1N protein of E. coli; (a) root mean square deviation (RMSD), (b) radius of gyration (Rg), (c) solvent accessible surface area, (d) hydrogen bond

Fig. 4.

Fig. 4

Open in a new tab

The molecular dynamics simulation analysis of (a) RMSF, and (b) MMPBSA binding free energies of complexes Glucosinolates and Indole-3-carbinol from Brassica oleracea with 2F1N protein of E. coli at 100 ns simulation period

The RMSD represents the mean displacement of a set of atoms relative to a reference structure within a specific frame (Sargsyan et al. 2017). In this investigation, we analyzed the RMSD of the Carbon (Cα) backbone values of the drug candidate compounds and a control (Ciprofloxacin) bound to CdtB protein (PDB ID: 2F1N) throughout a 100 ns simulation period as well as the apo protein. As can be seen from Fig. 3a, RMSD values of Glucosinolates complex with 2F1N maintained equilibrium from the initiation till the end of the evolution with minimum and maximum values of 0.396 Å to 1.943 Å. The average RMSD value observed was 1.565 Å, slightly higher than the control (1.271 Å) but aligned with the average RMSD of the apo-protein (1.486 Å). Likewise, the Cα atoms of Indole-3-carbinol with 2F1N complex maintained an equilibrium state throughout the 100 ns evolution time. In this case, the minimum and maximum RMSD values remained between 0.429 and 1.576 Å. The average RMSD value in this complex was 1.31 Å, which more closely mirrors the control and apo protein compared to Glucosinolates, indicating minimal conformational changes induced by ligand binding. The RMSD values for both ligands remained within narrow ranges, suggesting that the binding of these ligands did not induce significant conformational changes in the 2F1N protein backbone.

The assessment of the root mean square radial distance from the center of mass of the target protein to its terminals is referred to as the radius of gyration (Rg) (Arnittali et al. 2019). This measurement helps evaluate the mobility and rigidity of the protein in conjunction with specific ligands, offering insights into the compactness of the protein-ligand complex (Jana et al. 2018). Figure 3b displays the stability of ligand Glucosinolates and Indole-3-carbinol and the control (Ciprofloxacin) in complex with 2F1N along with apo protein, as determined by the Rg throughout a 100 ns simulation period. The ligand Glucosinolates exhibited a consistent trend from the initiation of the simulation until 60 ns, followed by fluctuations between 60 and 80 ns. Subsequently, it maintained a steady state until the end simulation, with recorded minimum, maximum, and average Rg values of 17.222 Å, 18.445 Å, and 17.551 Å, respectively. Despite the observed fluctuations, the average Rg value remained proximate to that of the apo-protein (17.229 Å) and control (17.387 Å). Figure 3b illustrates that Indole-3-carbinol demonstrated a more stable state throughout the simulation. The Rg values consistently ranged between 17.223 Å and 17.809 Å. The mean Rg value was calculated as 17.438 Å, which more closely aligns with the mean Rg value of the apo-protein and control.

Solvent accessible surface area (SASA) measures the contact area between a molecule and solvent, allowing comparison of different molecules or conformations, or measuring buried surface due to oligomerization (Ladanyi and Skaf 1993). Figure 3c shows the bimolecular surface area of 2F1N-ligand complexes, which is accessible to solvent molecules. The ligand Glucosinolates complexed with 2F1N exhibited a higher mean SASA value of 11774.704 Å2, with its SASA fluctuating between a maximum of 12362.373 Å2 and a minimum of 11173.662 Å2. The higher mean values compared to apo (11269.997 Å2) and control (11448.269 Å2) suggest considerable interaction with its surrounding solvent environment. Conversely, the 2F1N-Indole-3-carbinol complex displayed a comparably lower mean SASA of 11607.68 Å2, a value closer to both apo and control SASA mean, which may hint at a more constrained or rigid binding configuration. The trajectory analysis of this complex indicated minimal fluctuations, with SASA values ranging between 11051.84 and 12171.68 Å2, implying relatively stable interactions throughout the 100 ns simulation.

One of the most crucial interactions for maintaining the stability of the protein-ligand complex is hydrogen bonding (Ladanyi and Skaf 1993). The number of hydrogen bonds created in the 2F1N complexes throughout of 100 ns is displayed in Fig. 3d. The results showed that the control compound formed an average of 193.39 hydrogen bonds with 2F1N, indicating a strong interaction. Compared to the control, both ligands, Glucosinolates and Indole-3-carbinol, exhibited close hydrogen bonds with the target protein, indicating their potential efficacy as inhibitor molecules. Specifically, approximately 188.71 hydrogen bonds, on average, were formed in the 2F1N-Glucosinolates complex, while around 189.80 hydrogen bonds were formed in the 2F1N-Indole-3-carbinol complex.

The Root Mean Square Fluctuation (RMSF) values were plotted to assess residue-specific fluctuations in both the apo form (protein without a ligand) and its ligand-bound complexes (Pitera 2014). Deviations in RMSD for all forms reveal dynamic mobility, especially in loop regions (Sargsyan et al. 2017). Significant RMSF changes exceeding 3 Å are noteworthy, as they greatly affect amino acid flexibility (Pokhrel et al. 2021). Figure 4a presents the RMSF plot, highlighting rigid and flexible regions of the protein, serving as a validation measure for structural variability in ligand-protein complexes and demonstrating the role of specific residues in these changes. Amino acid deviations were calculated over a 100 ns simulation. For the apo form, significant deviations above 3 Å were observed in residues ALA19, ARG238, GLN240, SER18, and ARG162, while other residues fluctuated between 0.386 and 2.971 Å. The average RMSF for the apo form was 1.10 Å, and these deviations likely contribute to the stability seen in its RMSD over the simulation time. Comparing the RMSF values of the apo form with the ligand-bound complexes, the 2F1N-Glucosinolates complex showed higher deviations, with an average RMSD of 1.156 Å. Residues SER18, THR2, ARG80, ARG237, ARG162, ARG238, and GLY1 exceeded the 3 Å mark, contributing to the RMSD deviations between 50 and 75 ns. Similarly, the 2F1N + Indole-3-carbinol complex also exhibited fluctuations in residues ARG162, THR2, ARG238, and GLY1, with RMSF values above 3 Å. This complex had an average RMSF of 1.07 Å, lower than the apo form. The control complex of ciprofloxacin with 2F1N displayed the lowest average RMSF of 1.049 Å, with residues ARG237, THR2, ARG238, and GLY1 surpassing 3 Å. Higher RMSF values indicate greater protein flexibility, but overall, the protein-ligand complexes showed lower RMSF values, suggesting reduced flexibility in the bound state compared to the apo form.

MMPBSA

The prime MMPBSA integrated within the YASARA modelling software suite computed the docked complexes’ binding free energy (ΔG). The relative free binding energies of 2F1N + Glucosinolates and 2F1N + Indole-3-carbinol complexes were − 36.39 KJ/mol and − 20.00 KJ/mol, respectively, as shown in Fig. 4b. A negative MMPBSA binding energy suggests a favorable binding interaction. A ΔG value of approximately − 10 KJ/mol represents a solid binding and biologically significant interaction. The data indicates that the association between the ligands and the protein is stable.

ADMET analysis

The prediction of Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) properties are crucial in drug development to assess the pharmacokinetic behavior and potential toxicity of compounds (Ferreira and Andricopulo 2019; van de Waterbeemd and Gifford 2003). In this study, we utilized computational models to predict the ADMET properties of Glucosinolates and Indole-3-carbinol using SwissADME and Pkcsm server, focusing on parameters such as absorption, distribution, metabolism, excretion, and toxicity (Table 2). Absorption properties were evaluated based on water solubility, CaCO2 permeability and percentage of human intestinal absorption (HIA). Glucosinolates exhibited a water solubility of -2.683, while Indole-3-carbinol showed a water solubility of -1.874. These values are adequate for drug development compared to standard ranges where compounds with LogSw values less than (more negative than) -6 are considered poorly soluble (Maliehe et al. 2020). Glucosinolates show lower CaCO2 permeability (-0.848) than Indole-3-carbinol (1.586), with values >-5.0 indicating acceptable intestinal absorption for drug development (Wilson and Bell 1993). Moreover, the predicted HIA percentage was 5.082% for Glucosinolates and 92.085% for Indole-3-carbinol indicating their bioavailability. These values suggest differences in the absorption potential of the two compounds, with Indole-3-carbinol exhibiting higher intestinal absorption than Glucosinolates. Although the poor absorption of Glucosinolates can limit the therapeutic efficacy, higher doses or alternative formulations can be utilized to achieve sufficient systemic concentrations.

Table 2.

ADMET prediction glucosinolates and Indole-3-carbinol from SwissADME and PKCSM tools, where every compound had almost favorable drug-likeness properties

Parameters Molecules
Glucosinolates Indole-3-carbinol
Absorption Water solubility -2.683 -1.874
CaCO2 permeability -0.848 1.586
Intestinal absorption (human) (%) Low (5.082%) High (92.085%)
Distribution VDss (human) (log L/kg) -0.281 0.212
CNS permeability -4.048 -2.17
Metabolism CYP1A2 inhibitor No Yes
CYP2D6 inhibitor No No
P-glycoprotein I inhibitor No No
P-glycoprotein II inhibitor No No
Excretion Total Clearance 0.375 0.515
Renal OCT2 substrate No No
Toxicity AMES toxicity No No
Max. tolerated dose (human) (log mg/kg/day) 0.947 0.163
hERG I inhibitor No No
hERG II inhibitor No No
Oral Rat Acute Toxicity (LD50) (mol/kg) 2.537 2.39
Oral Rat Chronic Toxicity (LOAEL) (log mg/kg_bw/day) 3.72 1.77
Hepatotoxicity Yes No
Open in a new tab

The distribution of compounds within the body was assessed through parameters such as volume of distribution (VDss), and central nervous system (CNS) permeability (Kramer et al. 2018). Glucosinolates exhibit a low volume of distribution (VDss) in humans (-0.281 log L/kg) compared to Indole-3-carbinol (0.212 log L/kg), indicating restricted distribution within the body. Moreover, Glucosinolates showed significantly lower central nervous system (CNS) permeability (-4.048) compared to Indole-3-carbinol (-2.17), suggesting reduced potential for CNS penetration and are unlikely to cross the blood brain barrier. The metabolism of Glucosinolates and Indole-3-carbinol was evaluated in terms of their inhibition potential on essential cytochrome P450 (CYP) enzymes, including CYP1A2, CYP2D6, P-glycoprotein I and P-glycoprotein II. Glucosinolates exhibit no inhibition of CYP1A2 or CYP2D6 enzymes, suggesting minimal potential for metabolic interactions or toxicity concerns related to these pathways. This indicates a favorable metabolic profile for Glucosinolates, potentially reducing the risk of drug-drug interactions.

Moreover, Glucosinolates do not inhibit P-glycoprotein I or II, indicating no interference with drug efflux mechanisms. This suggests that Glucosinolates may not affect the absorption or distribution of co-administered drugs that are substrates of P-glycoprotein transporters, contributing to their favorable pharmacokinetic profile and reduced likelihood of pharmacokinetic interactions. Indole-3-carbinol, on the other hand, exhibits inhibition of CYP1A2 enzymes, suggesting a potential for metabolic interactions or toxicity concerns related to this pathway. This implies a need for careful consideration of potential drug-drug interactions when co-administering Indole-3-carbinol with drugs metabolized by CYP1A2. Like Glucosinolates, Indole-3-carbinol does not inhibit P-glycoprotein I or II, indicating no interference with drug efflux mechanisms. This suggests that Indole-3-carbinol may not affect the absorption or distribution of co-administered drugs that are substrates of P-glycoprotein transporters.

The total clearance values for Glucosinolates (0.375 log mL/min/kg) and Indole-3-carbinol (0.515 log mL/min/kg) suggest efficient elimination from the body, potentially reducing the risk of accumulation and associated toxicity. Additionally, neither compound is predicted to be a substrate for the renal OCT2 transport protein. Glucosinolates and Indole-3-carbinol demonstrate comparable toxicity profiles based on various parameters. Both compounds exhibit no AMES toxicity, indicating a lack of mutagenic potential. The maximum tolerated dose (MTD) for Glucosinolates (0.947 log mg/kg/day) is notably higher than that of Indole-3-carbinol (0.163 log mg/kg/day), suggesting differences in their tolerability in humans. Furthermore, neither compound inhibits hERG I or hERG II, indicating no potential for cardiac toxicity. Regarding acute toxicity in rats, Glucosinolates and Indole-3-carbinol show similar LD50 values (2.537 and 2.39 mol/kg, respectively), suggesting comparable acute toxicity profiles. However, Glucosinolates exhibit higher chronic toxicity (LOAEL) in rats (3.72 log mg/kg/bw/day) compared to Indole-3-carbinol (1.77 log mg/kg/bw/day), implying potential differences in long-term toxic effects Glucosinolates are predicted to be hepatotoxic, whereas Indole-3-carbinol is not associated with hepatotoxicity. While utilizing the advantages of indole-3-carbinol, it is possible to reduce the risks related to glucosinolates by monitoring on liver function and modifying dosages according to individual responses. Pharmacokinetic and Lipinski rule assessments were conducted for Glucosinolates and Indole-3-carbinol, aligning with established guidelines and in light of previous studies (Table 3) (Chen et al. 2020; Lipinski 2004; Pradeepkiran et al. 2021).

Table 3.

Lipinski rule and pharmacokinetics properties prediction glucosinolates and Indole-3-carbinol from SwissADME server where every compound had almost favorable drug-likeness properties

Pubchem ID 6,537,198 3712
Molecule name Glucosinolates Indole-3-carbinol
Formula C16H20N2O9S2 C9H9NO
Molecular weight 448.47 147.17
Number rotatable bonds 7 1
Number H-bond acceptors 10 1
Number H-bond donors 6 2
TPSA 215.58 36.02
Log Po/w (WLOGP) 0.46 1.51
Lipinski violations 2 violations: N or O > 10, NH or OH > 5 0
Bioavailability score 0.11 0.55
Open in a new tab

Glucosinolates, with a molecular weight of 448.47 Da, exhibit seven rotatable bonds, ten hydrogen bond acceptors, and six hydrogen bond donors, resulting in a total polar surface area (TPSA) of 215.58 Å2. While these values generally align with Lipinski’s Rule of Five, the compound violates the rule in two key parameters: the number of hydrogen bond donors (6) and hydrogen bond acceptors (10). According to Lipinski’s guidelines, drug-like compounds should have no more than 5 hydrogen bond donors and 10 hydrogen bond acceptors, however, as per this rule at least two violation is considered (Hosen et al. 2023; Rahman et al. 2024). Glucosinolates slightly exceed the donor limit, which could increase hydrophilicity and somewhat hinder membrane permeability. However, this effect is likely not significant. The acceptor count is at the upper limit but raises concerns about the compound’s ability to cross lipid membranes. As a natural phytochemical, Glucosinolates’ complex structure, featuring multiple functional groups and a high TPSA, leads to these violations. However, this complexity does not inherently indicate a lack of efficacy or therapeutic potential (Nguyen et al. 2020). Many natural products, including certain antibiotics, violate Lipinski’s rules but remain bioavailable or effective due to active transport mechanisms or other favorable pharmacokinetic properties. For instance, quinine, despite its high molecular weight and logP, has long been used to treat malaria (Lohohola et al. 2021). While Lipinski’s rules provide a useful framework for assessing drug-likeness, they are not absolute barriers. Given the pharmacokinetic and pharmacodynamic complexities, Glucosinolates may still be effective through oral or alternative administration routes (e.g., intravenous, topical).

Indole-3-carbinol, with a molecular weight of 147.17 Da, displayed one rotatable bond, one hydrogen bond acceptor, and two hydrogen bond donors, yielding a TPSA of 36.02 Å2, adhering to Lipinski’s rule of five. The lipophilicity values, represented by LogPo/w, ranged from 0.46 to 1.51 for Glucosinolates and 1.51 for Indole-3-carbinol, meeting the criteria for good human oral bioavailability (0 < logP < 3)(Qidwai, 2017). Furthermore, Glucosinolates exhibited a lower bioavailability score of 0.11 compared to Indole-3-carbinol’s score of 0.55, suggesting potential challenges in oral absorption for Glucosinolate (Table 3).

In vitro antibacterial activity

The in vitro antibacterial activity of Glucosinolates and Indole-3-carbinol against E. coli ATCC25922, a bacterium associated with GBC, was evaluated using the disc diffusion method. Both compounds exhibited varying degrees of inhibition against E. coli, as evidenced by the formation of clear zones around the impregnated discs on the agar plates (Fig. 5; Table 4). At the concentration of 25 µg/disc, Glucosinolates demonstrated a mean inhibition zone of 13.98 ± 0.153 mm diameters, which increased to 28.21 ± 0.501 mm and 34.25 ± 0.541 mm at concentrations of 50 µg/disc and 75 µg/disc, respectively (Fig. 5a, Fig. 6). Similarly, Indole-3-carbinol exhibited inhibitory effects with mean inhibition zone of 10.60 ± 0.702 mm, 25.57 ± 0.829 mm, and 28.67 ± 0.376 mm diameters at concentrations of 25 µg/disc, 50 µg/disc, and 75 µg/disc, respectively (Fig. 5b, Fig. 6). Comparatively, the standard antibiotic drug Ciprofloxacin (5 µg/disc) produced smaller inhibition zones of 23.30 ± 0.723 mm and 23.17 ± 0.441 mm, indicating lower activity than lead compounds. These results support previous studies where B. oleracea plant extracts are used as an antibacterial agent against E. coli (Akanmu et al. 2019; Alani et al. 2021; Zamir et al. 2013).

Fig. 5.

Fig. 5

Open in a new tab

Antibacterial activity of compound (a) Glucosinolates and (b) Indole-3-carbinol against E. coli at three different concentrations. Ciprofloxacin was used as a positive control

Table 4.

Zone of inhibition against E. Coli with glucosinolates and Indole-3-carbinol in different concentration. Ciprofloxacin is used as control

Compounds 25 µl/disc 50 µl/disc 75 µl/disc Control
Glucosinolates 13.98 ± 0.153 28.21 ± 0.501 34.25 ± 0.541 23.30 ± 0.723
Indole-3-carbinol 10.60 ± 0.702 25.57 ± 0.829 28.67 ± 0.376 23.17 ± 0.441
Open in a new tab

Fig. 6.

Fig. 6

Open in a new tab

Zone of inhibition against E. coli with Glucosinolates and Indole-3-carbinol in different concentrations. Ciprofloxacin is used as a control

Glucosinolates may contribute to antibiosis due to their antibacterial properties as evidenced by numerous prior studies (Aires et al. 2009a; b; Melrose 2019; Yadav et al. 2022). On the other side, the antibacterial efficacy of Indole-3-carbinol could influence bacterial inhibition, particularly against E. coli (Kwun et al. 2021; Liu et al. 2024; Monte et al. 2014). Glucosinolates, upon undergoing hydrolysis, facilitated by the enzyme myrosinase, generate distinct compounds that exert influence on the proliferation of various microorganisms (Favela-González et al. 2020). Glucosinolates can also retard the growth of several pathogenic bacteria biofilm formation and cytotoxic effect against E. coli (Vig et al. 2009). Indole-3-carbinol can hinder efflux pumps, biofilm formation, FtsZ activity, and MRSA pyruvate kinase activity of bacteria (Liu et al. 2020). Based on the results discussed above, Glucosinolates and Indole-3-carbinol emerge as valuable blueprints for creating innovative antibacterial agents, showing great potential for further exploration and utilization in future research and practical applications.

DFT analysis

Frontier molecular orbitals (FMOs) are collectively called highest-occupied (HOMOs) and lowest-lying unoccupied (LUMOs) molecular orbitals. FMOs are important in determining the optical and electrical characteristics of molecules. A molecule’s ability to donate electrons is represented by its HOMO energy, and its ability to accept electrons is represented by its LUMO energy (Rasouli et al. 2017). The kinetic stability and reactivity of the molecule are controlled by the energy gap between HOMO and LUMO, where a bigger gap denotes higher stability (Alghamdi et al. 2023). The HOMO-LUMO gap is crucial for charge transfer interactions, influencing a molecule’s biological activity and spectroscopic properties, as HOMO-1 and LUMO + 1 may participate in reactions (Thiruvangoth 2024). Molecular orbitals of Glucosinolates and Indole-3-carbinol were generated and visualized using DFT shown in Fig. 7.

Fig. 7.

Fig. 7

Open in a new tab

Molecular orbitals of HOMO and LUMO of Glucosinolates and Indole-3-carbinol

The DFT (Density Functional Theory) results reveal significant variations in the electronic structures of glucosinolates and indole-3-carbinol, which directly affect their interactions with active sites (Rasouli et al. 2017). In glucosinolates, the HOMO (EH= -5.43 eV) is delocalized over the sulfur atom and aromatic ring, making these regions prone to nucleophilic attacks, while the LUMO (EL= -0.93 eV) is also spread over the aromatic ring, indicating susceptibility to electrophilic interactions (Fig. 7). The relatively small HOMO-LUMO gap (4.50 eV) suggests higher reactivity, enabling glucosinolates to engage in rapid electron transfers at active sites, such as those found in enzyme catalysis. In contrast, indole-3-carbinol has its HOMO (EH= -5.54 eV) localized on the indole ring, particularly near the nitrogen atom, suggesting nucleophilic tendencies in these areas. The LUMO (EH= -0.40 eV) is similarly focused on the ring, with a larger HOMO-LUMO gap (5.14 eV) (Fig. 7), implying lower reactivity and greater selectivity in its interactions, making it more stable and suited for targeted interactions at specific active site residues (Rasouli et al., 2022).

The variation in the molecular orbital distribution between the two compounds affects their nucleophilic and electrophilic behavior, reactivity, and selectivity (Rasouli et al., 2022). Glucosinolates, with a smaller energy gap, are more reactive but less selective, engaging with a wider range of active sites due to the delocalized nature of their orbitals. Indole-3-carbinol, with a larger HOMO-LUMO gap and more localized orbitals, exhibits greater selectivity and forms more specific interactions, particularly at nucleophilic or electrophilic centers. These differences in electronic properties highlight how glucosinolates and indole-3-carbinol may influence molecular interactions differently, providing crucial insights for designing ligands tailored to specific biochemical pathways or therapeutic targets.

The overall study lies in its exploration of naturally derived phytochemicals from broccoli (Brassica oleracea), specifically Glucosinolates and Indole-3-carbinol, as potential antibacterial agents against E. coli (ATCC25922). By combining both in vitro and in silico approaches, this research provides a comprehensive analysis of the antibacterial properties and molecular interactions of these compounds, targeting the E. coli CdtB protein. This dual approach not only validates the antibacterial effects empirically but also offers molecular-level insights that can guide the development of new, targeted antibacterial therapies.

Conclusion

In Conclusion, this study investigated the inhibitory potential of phytochemicals derived from Brassica oleracea against the CdtB protein (PDB ID: 2F1N) of Escherichia coli, a bacterium implicated in gallbladder cancer (GBC). Through molecular docking simulation, Glucosinolates and Indole-3-carbinol emerged as promising inhibitors, demonstrating robust binding affinities to the active sites of the target proteins. Molecular dynamics simulations further confirmed the stability of these protein-ligand complexes over 100 nanoseconds, suggesting their potential as lead compounds for drug development. Moreover, MMGBSA calculations revealed the favorable binding free energies of the docked complexes. In vitro antibacterial assays revealed the inhibitory effects of Glucosinolates and Indole-3-carbinol against E. coli, supporting their candidacy as antibacterial agents. ADMET profiling highlighted differences in absorption, distribution, metabolism, excretion, and toxicity between the two compounds, offering pertinent details for further optimization. The DFT studies indicated that Indole-3-carbinol is more chemically stable but less reactive than Glucosinolates. Nevertheless, conducting thorough clinical trials and translational studies is imperative to validate these discoveries rigorously. These results provide optimism for developing novel approaches to combat E. coli, a bacterium associated with GBC, thereby tackling a pressing concern in modern medicine.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1 (63.1MB, rar)

Acknowledgements

The authors want to thanks State Key Laboratory of Microbiology and Bioinformatics, Department of Microbiology, Shaheed Shamsuzzoha Institute of Biosciences, Affiliated with University of Rajshahi, Rajshahi, Bangladesh.

Author contributions

Md. Eram Hosen and Faria Tasnim: Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Methodology, Validation, Project administration, Resources, Supervision, writing – original draft, Writing – review & editing. Md. Enamul Haque, Ariful Islam, Mst Naharina Nuryay, Delara Yesmin, Jannatul Mawya, Najnin Akter, Nilima Rahman, Md Mosabbir Hossain, B.M. Mahmudul Hasan, Md. Naimul Hassan, Md Mahmudul Islam, Md. Khalekuzzaman: Data curation, Formal analysis, Methodology, Visualization, Validation, writing – original draft, Writing – review & editing.

Funding

The authors received no funding from an external source.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval

Not applicable.

Informed consent

Not applicable.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Co-First author: Faria Tasnim and Md. Eram Hosen.

References

  1. Aires A, Mota VR, Saavedra MJ, Rosa EAS, Bennett RN (2009) The antimicrobial effects of glucosinolates and their respective enzymatic hydrolysis products on bacteria isolated from the human intestinal tract. J Appl Microbiol 106(6):2086–2095. 10.1111/j.1365-2672.2009.04180.x [DOI] [PubMed] [Google Scholar]
  2. Aires A, Mota VR, Saavedra MJ, Monteiro AA, Simões M, Rosa EAS, Bennett RN (2009a) Initial in vitro evaluations of the antibacterial activities of glucosinolate enzymatic hydrolysis products against plant pathogenic bacteria. J Appl Microbiol 106(6):2096–2105. 10.1111/j.1365-2672.2009.04181.x [DOI] [PubMed] [Google Scholar]
  3. Akanmu AO, Balogun ST, Haruna F, Gamache S (2019) Antibacterial activities of leaf extracts of Brassica Oleracae var. capitata. (Brassicaceae) against multi-drug resistant clinical isolates in Maiduguri. Sokoto J Vet Sci 16(4):58. 10.4314/sokjvs.v16i4.7 [Google Scholar]
  4. Alani H, Ozaslan M, Karagoz ID, Simitcioglu B (2021) Investigation of Antimicrobial, DNA protective and cytotoxic activity of Red Cabbage (Brassica Oleracea L. Var. Capitata F. Rubra) Plant. Eurasia Proc Sci Technol Eng Math 12:123–129. 10.55549/epstem.1038398 [Google Scholar]
  5. Alghamdi SK, Abbas F, Hussein RK, Alhamzani AG, El-Shamy NT (2023) Spectroscopic characterization (IR, UV-Vis), and HOMO-LUMO MEP, NLO, NBO analysis and the antifungal activity for 4-Bromo-N-(2-nitrophenyl) benzamide; Using DFT modeling and in silico molecular docking. J Mol Struct 1271:134 [Google Scholar]
  6. Arnittali M, Rissanou AN, Harmandaris V (2019) Structure of biomolecules through molecular dynamics simulations. Proced Comput Sci 156:69–78. 10.1016/j.procs.2019.08.181 [Google Scholar]
  7. Baildya N, Khan AA, Ghosh NN, Dutta T, Chattopadhyay AP (2021) Screening of potential drug from Azadirachta Indica (neem) extracts for SARS-CoV-2: an insight from molecular docking and MD-simulation studies. J Mol Struct 1227:129390 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bateman NW, Conrads TP (2018) Recent advances and opportunities in proteomic analyses of tumour heterogeneity. J Pathol 244(5):628–637. 10.1002/path.5036 [DOI] [PubMed] [Google Scholar]
  9. Bendary MM, Abdel-Hamid MI, Alshareef WA, Alshareef HM, Mosbah RA, Omar NN, Al-Sanea MM, Alhomrani M, Alamri AS, Moustafa WH (2022) Comparative analysis of human and animal E. coli: serotyping, antimicrobial resistance, and virulence gene profiling. Antibiotics 11(5):552 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bianchi G, Picchi V, Tava A, Doria F, Walley PG, Dever L, di Bella MC, Arena D, Ammar H, Ben, Scalzo R, Lo (2024) Insights into the phytochemical composition of selected genotypes of organic kale (Brassica oleracea L. var. Acephala). J Food Compos Anal 125:105721 [Google Scholar]
  11. Bitew A, Tsige E (2020) High prevalence of Multidrug-resistant and extended‐spectrum β‐Lactamase‐producing Enterobacteriaceae: a cross‐sectional study at Arsho Advanced Medical Laboratory, Addis Ababa, Ethiopia. J Trop Med 2020(1):6167234 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Burley SK, Berman HM, Kleywegt GJ, Markley JL, Nakamura H, Velankar S (2017) Protein Data Bank (PDB): the single global Macromolecular structure archive. 627–641. 10.1007/978-1-4939-7000-1_26 [DOI] [PMC free article] [PubMed]
  13. Chen X, Li H, Tian L, Li Q, Luo J, Zhang Y (2020) Analysis of the physicochemical properties of acaricides based on Lipinski’s rule of five. J Comput Biol 27(9):1397–1406 [DOI] [PubMed] [Google Scholar]
  14. Cock I, Mavuso N, Van Vuuren S (2021) A review of plant-based therapies for the treatment of urinary tract infections in traditional southern African Medicine. Evidence‐Based Compl Altern Med 2021(1):7341124 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Daina A, Michielin O, Zoete V (2017) SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep 7(1):1–13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. de Ruyck J, Brysbaert G, Blossey R, Lensink M (2016) Molecular docking as a popular tool in drug design, an in silico travel. Adv Appl Bioinf Chem 9:1–11. 10.2147/AABC.S105289 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Durrant JD, McCammon JA (2011) Molecular dynamics simulations and drug discovery. BMC Biol 9(1):71. 10.1186/1741-7007-9-71 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Eissa M (2024) Escherichia coli: epidemiology, impact, antimicrobial resistance and prevention: a review. J Public Health Commun Med 1(1):39. 10.5455/JPHCM.20240110064652 [Google Scholar]
  19. Emad E, El-Katori, Ashraf S, Abousalem (2019) Inhibitive properties and computational approach of organoselenides on mild steel corrosion in acidic environment. Russ J Electrochem 55(12):1320–1335. 10.1134/S1023193519120048 [Google Scholar]
  20. Essmann U, Perera L, Berkowitz ML, Darden T, Lee H, Pedersen LG (1995) A smooth particle mesh Ewald method. J Chem Phys 103(19):8577–8593 [Google Scholar]
  21. Eyvazi S, Vostakolaei MA, Dilmaghani A, Borumandi O, Hejazi MS, Kahroba H, Tarhriz V (2020) The oncogenic roles of bacterial infections in development of cancer. Microb Pathog 141:104019. 10.1016/j.micpath.2020.104019 [DOI] [PubMed] [Google Scholar]
  22. Favela-González KM, Hernández‐Almanza AY, De la Fuente‐Salcido NM (2020) The value of bioactive compounds of cruciferous vegetables (Brassica) as antimicrobials and antioxidants: a review. J Food Biochem 44(10). 10.1111/jfbc.13414 [DOI] [PubMed]
  23. Ferreira LLG, Andricopulo AD (2019) ADMET modeling approaches in drug discovery. Drug Discov Today 24(5):1157–1165. 10.1016/j.drudis.2019.03.015 [DOI] [PubMed] [Google Scholar]
  24. Ferreira L, Dos Santos R, Oliva G, Andricopulo A (2015) Molecular Docking and structure-based drug design strategies. Molecules 20(7):13384–13421. 10.3390/molecules200713384 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Foudraine DE, Strepis N, Stingl C, Ten Kate MT, Verbon A, Klaassen CHW, Goessens WHF, Luider TM, Dekker LJM (2021) Exploring antimicrobial resistance to beta-lactams, aminoglycosides and fluoroquinolones in E. Coli and K. pneumoniae using proteogenomics. Sci Rep 11(1):12472 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Guex N, Peitsch MC (1997) SWISS-MODEL and the Swiss‐pdb viewer: an environment for comparative protein modeling. Electrophoresis 18(15):2714–2723 [DOI] [PubMed] [Google Scholar]
  27. Hanwell MD, Curtis DE, Lonie DC, Vandermeersch T, Zurek E, Hutchison GR (2012) Avogadro: an advanced semantic chemical editor, visualization, and analysis platform. J Cheminform 4(1):1–17 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Harrach MF, Drossel B (2014) Structure and dynamics of TIP3P, TIP4P, and TIP5P water near smooth and atomistic walls of different hydroaffinity. J Chem Phys 140(17):174501 [DOI] [PubMed] [Google Scholar]
  29. Hollingsworth SA, Dror RO (2018) Molecular dynamics simulation for all. Neuron 99(6):1129–1143 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Hosen ME, Supti J, Akash S, Rahman S, Faruqe ME, Manirujjaman MO, Acharjee M, Gaafar UK, Ouahmane A-RZ, Sitotaw L, Bourhia B, M., Zaman R (2023) Mechanistic insight of Staphylococcus aureus associated skin cancer in humans by Santalum album derived phytochemicals: an extensive computational and experimental approaches. Front Chem 11. 10.3389/fchem.2023.1273408 [DOI] [PMC free article] [PubMed]
  31. Hynes D, Duffin C, Catanzano T (2020) Infections of the hepatobiliary system. Semin Ultrasound CT MRI 41(1):46–62 [DOI] [PubMed] [Google Scholar]
  32. Inc AS (2012) Discovery studio modeling environment, release 3.5. Accelrys Discovery Studio. Accelrys Software Inc, San Diego
  33. Islam R, Parves MR, Paul AS, Uddin N, Rahman MS, Mamun A, Al, Hossain MN, Ali MA, Halim MA (2020) A molecular modeling approach to identify effective antiviral phytochemicals against the main protease of SARS-CoV-2. J Biomol Struct Dyn 1–12. 10.1080/07391102.2020.1761883 [DOI] [PMC free article] [PubMed]
  34. Jana S, Ganeshpurkar A, Singh SK (2018) Multiple 3D-QSAR modeling, e-pharmacophore, molecular docking, and in vitro study to explore novel AChE inhibitors. RSC Adv 8(69):39477–39495. 10.1039/C8RA08198K [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Kandsi F, Elbouzidi A, Lafdil FZ, Meskali N, Azghar A, Addi M, Hano C, Maleb A, Gseyra N (2022) Antibacterial and antioxidant activity of Dysphania ambrosioides (L.) Mosyakin and Clemants essential oils: experimental and computational approaches. Antibiotics 11(4):482. 10.3390/antibiotics11040482 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Kramer C, Ting A, Zheng H, Hert J, Schindler T, Stahl M, Robb G, Crawford JJ, Blaney J, Montague S, Leach AG, Dossetter AG, Griffen EJ (2018) Learning medicinal chemistry absorption, distribution, metabolism, excretion, and toxicity (ADMET) rules from cross-company matched molecular pairs analysis (MMPA). J Med Chem 61(8):3277–3292. 10.1021/acs.jmedchem.7b00935 [DOI] [PubMed] [Google Scholar]
  37. Krieger E, Vriend G (2015) New ways to boost molecular dynamics simulations. J Comput Chem 36(13):996–1007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Krieger E, Dunbrack RL, Hooft RWW, Krieger B (2012) Assignment of protonation states in proteins and ligands: combining pK a prediction with hydrogen bonding network optimization. Computational drug discovery and design. Springer, pp 405–421 [DOI] [PubMed] [Google Scholar]
  39. Kumar Deb P, Shilkar D, Sarkar B (2022) UHPLC-ESI‐QTOF‐MS/MS Based identification, quantification, and assessment of in silico molecular interactions of major phytochemicals from bioactive fractions of Clerodendrum Glandulosum Lindl. Leaves. Chem Biodivers 19(10). 10.1002/cbdv.202200617 [DOI] [PubMed]
  40. Kwun MS, Yun JE, Lee DG (2021) Indole-3-carbinol induces apoptosis-like death in Escherichia coli on different contribution of respective reactive oxygen species. Life Sci, 275. 10.1016/j.lfs.2021.119361 [DOI] [PubMed]
  41. Ladanyi BM, Skaf MS (1993) Computer simulation of hydrogen-bonding liquids. Annu Rev Phys Chem 44(1):335–368. 10.1146/annurev.pc.44.100193.002003 [Google Scholar]
  42. Land H, Humble MS (2018) YASARA: a tool to obtain structural guidance in biocatalytic investigations. Protein Engineering. Springer, pp 43–67 [DOI] [PubMed] [Google Scholar]
  43. Li H, Xia Y, Liu H-Y, Guo H, He X-Q, Liu Y, Wu D-T, Mai Y-H, Li H-B, Zou L, Gan R-Y (2022) Nutritional values, beneficial effects, and food applications of broccoli (Brassica oleracea var. italica Plenck). Trends Food Sci Technol 119:288–308. 10.1016/j.tifs.2021.12.015 [Google Scholar]
  44. Lipinski CA (2004) Lead- and drug-like compounds: the rule-of-five revolution. Drug Discov Today Technol 1(4):337–341. 10.1016/j.ddtec.2004.11.007 [DOI] [PubMed] [Google Scholar]
  45. Liu Y, Cui Y, Lu L, Gong Y, Han W, Piao G (2020) Natural indole-containing alkaloids and their antibacterial activities. Arch Pharm 353(10). 10.1002/ardp.202000120 [DOI] [PubMed]
  46. Liu S, Li W, Chen J, Li M, Geng Y, Liu Y, Wu W (2024) The footprint of gut microbiota in gallbladder cancer: a mechanistic review. Front Cell Infect Microbiol 14. 10.3389/fcimb.2024.1374238 [DOI] [PMC free article] [PubMed]
  47. Lohohola PO, Mbala BM, Bambi S-MN, Mawete DT, Matondo A, Mvondo JGM (2021) In silico ADME/T properties of quinine derivatives using SwissADME and pkCSM webservers. Int J Trop DISEASE Health 42(11):1–12 [Google Scholar]
  48. Maliehe TS, Tsilo PH, Shandu JS (2020) Computational evaluation of ADMET properties and Bioactive score of compounds from Encephalartos ferox. Pharmacogn J 12(6):1357–1362. 10.5530/pj.2020.12.187 [Google Scholar]
  49. Martinson JNV, Walk ST (2020) Escherichia coli residency in the gut of healthy human adults. EcoSal Plus 9(1):10–1128 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Melrose J (2019) The glucosinolates: a sulphur glucoside family of mustard anti-tumour and antimicrobial phytochemicals of potential therapeutic application. Biomedicines 7(3):62. 10.3390/biomedicines7030062 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Monte J, Abreu AC, Borges A, Simões LC, Simões M (2014) Antimicrobial activity of selected phytochemicals against Escherichia coli and Staphylococcus aureus and their biofilms. Pathogens 3(2):473–498. 10.3390/pathogens3020473 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Nath G (2010) Role of bacteria in carcinogenesis, with special reference to carcinoma of the gallbladder. World J Gastroenterol 16(43):5395. 10.3748/wjg.v16.i43.5395 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Nguyen VPT, Stewart J, Lopez M, Ioannou I, Allais F (2020) Glucosinolates: natural occurrence, biosynthesis, accessibility, isolation, structures, and biological activities. Molecules 25(19):4537 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Nougayrède J-P, Chagneau CV, Motta J-P, Bossuet-Greif N, Belloy M, Taieb F, Gratadoux J-J, Thomas M, Langella P, Oswald E (2021) A toxic friend: genotoxic and mutagenic activity of the probiotic strain Escherichia coli Nissle 1917. MSphere 6(4):10–1128 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Peng Z, Wang X, Huang J, Li B (2024) Pathogenic Escherichia coli. Molecular Medical Microbiology. Elsevier, pp 1065–1096 [Google Scholar]
  56. Pires DE, V, Blundell TL, Ascher DB (2015) pkCSM: predicting small-molecule pharmacokinetic and toxicity properties using graph-based signatures. J Med Chem 58(9):4066–4072 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Pitera JW (2014) Expected distributions of root-mean-square positional deviations in proteins. J Phys Chem B 118(24):6526–6530 [DOI] [PubMed] [Google Scholar]
  58. Pokharel P, Dhakal S, Dozois CM (2023) The diversity of Escherichia coli pathotypes and vaccination strategies against this versatile bacterial pathogen. Microorganisms 11(2):344 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Pokhrel S, Bouback TA, Samad A, Nur SM, Alam R, Abdullah-Al-Mamun M, Nain Z, Imon RR, Talukder MEK, Tareq MMI, Hossen MS, Karpiński TM, Ahammad F, Qadri I, Rahman MS (2021) Spike protein recognizer receptor ACE2 targeted identification of potential natural antiviral drug candidates against SARS-CoV-2. Int J Biol Macromol 191:1114–1125. 10.1016/j.ijbiomac.2021.09.146 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Pollastri MP (2010) Overview on the rule of five. Curr Protocols Pharmacol 49(1). 10.1002/0471141755.ph0912s49 [DOI] [PubMed]
  61. Pons BJ, Pettes-Duler A, Naylies C, Taieb F, Bouchenot C, Hashim S, Rouimi P, Deslande M, Lippi Y, Mirey G, Vignard J (2021) Chronic exposure to Cytolethal Distending Toxin (CDT) promotes a cGAS-dependent type I interferon response. Cell Mol Life Sci 78(17–18):6319–6335. 10.1007/s00018-021-03902-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Pradeepkiran JA, Sainath SB, Shrikanya KVL (2021) In silico validation and ADMET analysis for the best lead molecules. In Brucella Melitensis, Elsevier, pp 133–176. 10.1016/B978-0-323-85681-2.00008-2
  63. Rahman MS, Hosen ME, Faruqe MO, Khalekuzzaman M, Islam MA, Acharjee UK, Bin Jardan YA, Nafidi H-A, Mekonnen AB, Bourhia M (2024) Evaluation of Adenanthera pavonina-derived compounds against diabetes mellitus: insight into the phytochemical analysis and in silico assays. Front Mol Biosci 10:1278701 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Rasouli H, Hosseini-Ghazvini SM-B, Adibi H, Khodarahmi R (2017) Differential α-amylase/α-glucosidase inhibitory activities of plant-derived phenolic compounds: a virtual screening perspective for the treatment of obesity and diabetes. Food Funct 8(5):1942–1954 [DOI] [PubMed] [Google Scholar]
  65. Rasouli H, Hosseini Ghazvini SMB, Yarani R, Altıntaş A, Jooneghani SGN, Ramalho TC (2022) Deciphering inhibitory activity of flavonoids against tau protein kinases: a coupled molecular docking and quantum chemical study. J Biomol Struct Dyn 40(1):411–424 [DOI] [PubMed] [Google Scholar]
  66. Salomon-Ferrer R, Case DA, Walker RC (2013) An overview of the Amber biomolecular simulation package. WIREs Comput Mol Sci 3(2):198–210. 10.1002/wcms.1121 [Google Scholar]
  67. Sargsyan K, Grauffel C, Lim C (2017) How molecular size impacts RMSD applications in molecular dynamics simulations. J Chem Theory Comput 13(4):1518–1524. 10.1021/acs.jctc.7b00028 [DOI] [PubMed] [Google Scholar]
  68. Syed RU, Moni SS, Break MK, Bin, Khojali WMA, Jafar M, Alshammari MD, Abdelsalam K, Taymour S, Alreshidi KSM, Taha E, Mohan S (2023) Broccoli: a multi-faceted vegetable for health: an in-depth review of its nutritional attributes, antimicrobial abilities, and anti-inflammatory properties. Antibiotics 12(7):1157. 10.3390/antibiotics12071157 [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Thiruvangoth S (2024) Exploring the spectral and nonlinear optical properties of phenyl isothiocyanate through density functional theory and molecular docking studies: a comprehensive analysis of a potent biologically active phytochemical. Hybrid Adv 6:100214. 10.1016/j.hybadv.2024.100214 [Google Scholar]
  70. Tiruneh TA, Tiruneh GA, Abebe EC, Ayele TM (2022) Phytochemical Investigation and determination of antibacterial activity of Solvent leave extracts of Carissa Spinarum. Infect Drug Resist 15:807 [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Trott O, Olson AJ (2010) AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem 31(2):455–461. 10.1002/jcc.21334 [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Tsuchiya Y, Loza E, Villa-Gomez G, Trujillo CC, Baez S, Asai T, Ikoma T, Endoh K, Nakamura K (2018) Metagenomics of microbial communities in gallbladder bile from patients with gallbladder cancer or cholelithiasis. Asian Pac J Cancer Prev 19(4):961–967. 10.22034/APJCP.2018.19.4.961 [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. van de Waterbeemd H, Gifford E (2003) ADMET in silico modelling: Towards prediction paradise? Nat Rev Drug Discovery 2(3):192–204. 10.1038/nrd1032 [DOI] [PubMed] [Google Scholar]
  74. Vig AP, Rampal G, Thind TS, Arora S (2009) Bio-protective effects of glucosinolates – a review. LWT Food Sci Technol 42(10):1561–1572. 10.1016/j.lwt.2009.05.023 [Google Scholar]
  75. Wang H, Gong J, Chen J, Zhang W, Sun Y, Sun D (2024) Intestinal microbiota and biliary system diseases. Front Cell Infect Microbiol 14:1362933 [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Wilson R, Bell MV (1993) Molecular species composition of glycerophospholipids from white matter of human brain. Lipids 28(1):13–17 [DOI] [PubMed] [Google Scholar]
  77. Yadav S, Teng P-Y, Choi J, Singh AK, Vaddu S, Thippareddi H, Kim WK (2022) Influence of rapeseed, canola meal and glucosinolate metabolite (AITC) as potential antimicrobials: effects on growth performance, and gut health in Salmonella Typhimurium challenged broiler chickens. Poult Sci 101(1):101551. 10.1016/j.psj.2021.101551 [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Zamir T, Farooqui R, Rajput MA, Mustafa K (2013) In-vitro assessment of antibacterial activity of methanol extract of Brassica oleracae against selected bacterias. J Liaquat Univ Med Health Sci 12(3):177–181 [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1 (63.1MB, rar)

Data Availability Statement

No datasets were generated or analysed during the current study.

Articles from In Silico Pharmacology are provided here courtesy of Springer

RESOURCES