In silico investigation of green tea (Camellia sinensis) phytochemicals as potential inhibitors of Haemonchus contortus cystatin-1 through Homology Modeling and Molecular Docking

Introduction

Haemonchus contortus (Barber's pole worm) is prevalent globally that affects the small ruminants. The parasite resides in the abomasum of sheep and goats, sucks the blood causing anaemia, reduction in weight gain, decreased productivity, and in severe cases leading to death, makes the disease one of the most economically significant parasitic diseases [1] . Emergence of resistance to conventional anthelmintic drugs have become major challenges to sustainable livestock production. Consequently, the search for novel, effective, and environmentally safe antiparasitic agents has gained increasing attention. The proteins of H. contortus mediate several important functions of the parasites, such as hostparasite interactions, nutrient absorption, immune evasion, reproduction, their survival and development, and cause pathogenecity. Several enzymes like cysteine proteases, glutathione transferases, and somatic proteins, excretory-secretory proteins are identified as potential therapeutic targets. Identification and functional characterization of the proteins as drug targets are corroborated by the advancement in genomics, proteomics, transcriptomic studies of H. contortus [2] . The identification and optimization of bioactive molecules against parasite diseases has been made possible by computer-aided drug discovery (CADD). Techniques such as molecular docking, molecular dynamics simulations, virtual screening, and ADMET (absorption, distribution, metabolism, excretion, and toxicity) prediction enable rapid evaluation of ligandtarget interactions while reducing the time and cost associated with conventional drug development [3] . Prior to experimental validation, these in silico methods offer important insights into the pharmacokinetic characteristics, binding affinities, and interaction mechanisms of possible drug candidates.

Natural products continue to serve as an important source of antiparasitic compounds. Camellia sinensis (green tea) has attracted considerable interest due to their diverse biological activities such as antioxidant, antimicrobial, antiinflammatory, and antiparasitic effects. The green tea is a rich source of bioactive phytochemicals [4] . Previous study described the in vitro anthelmintic potential of green tea [5] . Several studies identified that catechins of green tea possess anthelmintic potential affecting the biotic activity of the helminths [6] . The present study explores the anthelmintic activity of phytochemicals of green tea against cystatin-1 of H. contortus. The findings may contribute to the development of novel plant-based therapeutic agents for the control of haemonchosis and help address the growing problem of anthelmintic resistance.

Materials and methods

Prediction of three dimensional structure of H. contortus cystatin-1:

The amino acid sequence of the cystatin-1 with 122 amino acid residues were obtained from NCBI database (Accession number: AAB95324.1). Three dimensional model of cystatin-1 was predicted in SWISS-MODEL server (https://swissmodel.expasy.org/) through automated homology modeling by submitting the retrieved amino acid sequence. The most appropriate structural template for the construction of protein three dimensional model was identified on the basis of sequence similarity, and favorable GMQE score [7] .

Validation and refinement of predicted three dimensional model of H. contortus cystatin-1

The predicted cystatin-1 model was validated using ERRAT, and PROCHECK program of SAVESv6.1 server. GalaxyRefine (https://galaxy.seoklab.org/cgi-bin/submit.cgi?type=REFINE) server was used to refine the protein model [8] . This was followed by the validation of the refined model in SAVESv6.1 server [9] .

Identification and selection of ligands for the molecular docking study: Phytochemicals present in Camellia sinensis (green tea) were collected and compiled from previously published literature report [4,10] . The SMILES representations of the identified phytochemicals of green tea were retrieved from PubChem database and submitted to SwissADME server. The server predicts the pharmacokinetic properties, physiochemical properties, lipophilicity, solubility, and druglikeness of the compounds. Based on the physiochemical properties, the drug-likeness of the compounds were determined using lipinski, veber, and egan's rule as filters. Lipophilicity, solubility, and pharamacokinetic properties i.e. absorption, distribution, metabolism, and excretion, druglikeness, and bioavailability score of the compounds were considered for the selection of the compounds for further molecular docking study. The toxicity of the phytochemicals were predicted in ProTox-III server (https://tox.charite.de/protox3/index.php?site=home [11] .

Preparation of H. contortus cystatin-1 and ligands for molecular docking study: Preparation of the refined H. contortus cystatin-1 model was done in AutoDock tools by checking for missing atoms, polar hydrogen atom addition, and assigning AutoDock4 (AD4) atom type to the protein model.

Energy minimization of the ligands was performed using Merck Molecular Force Field 1994 (MMFF94) and minimization was run up to 250 steps. The SDF file of each ligand was converted into PDB file in OpenBabel tool. Ligands were then prepared in AutoDock tools by addition by polar hydrogens and gasteiger charges.

Molecular docking and interaction analysis

The prepared protein model and ligands were docked by placing them in the grid box. The grid box was set in a way to cover the entire protein surface to allow the ligand to bind to the binding pocket of the protein. Autogrid and Autodock programs were run to perform the blind docking of the H. contortus cystatin-1, and selected ligands. The binding energy (kcal/mol) of the protein-ligand binding and the amino acids involved in the interaction, type of interaction and their bond length were analyzed to find the binding stability of the ligands within the binding pocket of the protein. The interaction of the protein and ligand were analyzed in a web tool, PLIP (Protein -Ligand Interaction Profiler) [12] (https://plip-tool.biotec.tu-dresden.de/plip-web/plip/index). The protein-ligand complex was visualized using Discovery Studio Visualizer [13] in 2D and 3D formats and images of the complex were generated.

Results and Discussion

In SWISS-MODEL server, cystatin domain-containing protein Alphafold DB model of AOA714YZ14_Haeco, H. contortus was identified as template sequence based on the sequence identity of 94.26%, and GMQE score of 0.93 to generate H. contortus cystatin-1 model (Fig 1). The GMQE score expressed between 0 and 1, higher number indicates the very high expected quality of the model [7] . Validation of the predicted H. contortus cystatin-1 model using SAVESv6.1 server resulted in an ERRAT score of 81.25, and Ramachandran plot generated by PROCHECK exhibited 92.4% residues in most favored region, 6.7% residues in allowed region, 1% residues in the generously allowed region, and 0% residues in the disallowed region (Fig 2)

. More than 90% residues in the most favored region, and no residues in the disallowed region indicated the good stereochemical quality of the model [14] . Generally, the ERRAT score of the three dimensional model more than 90, considered as the highest quality model [15] . Therefore, the predicted H. contortus cystatin-1 model, was refined in GalaxyRefine server to improve the quality and reliability of the protein model before the downstream application (Fig 3). Structural validation by the PROCHECK-generated Ramachandran plot revealed that 92.4% of residues were located in the most favored regions, 6.7% in the additionally allowed regions, 1.0% in the generously allowed regions, and 0% in the disallowed regions.

Validation of the refined model, yielded an ERRAT score of 92, and 95.2% residues were found in the most favorable region in Ramanchandran plot. The refinement of the protein model had enhanced the non-bonded atomic interactions, and reduced the structural errors, and few of the residues from the allowed region were redistributed into the most favored region (Figure 4). This process improved the overall geometric and stereochemical quality of H. contortus cystatin-1 model, made this protein model suitable for the molecular docking study. A total of 49 phytochemicals compounds were identified from the literature study (Table 1). SMILES of the compounds that were available in the PubChem database were obtained and submitted to SwissADME server.

Phytochemical compounds were selected for the molecular docking study based on the predicted lipophilicity score, solubility class, gastrointestinal absorption, cytochrome P450 inhibitors, lipinski's rule, veber's rule, and egan's rule, and bioavailability score of the compounds. Lipophilicity score (LogP) of the compounds should be between 0 and 5, moderately soluble to highly soluble with high gastrointestinal absorption, bioavailability score of 0.5 or more than 0.5, non-inhibitor of all five CYP450s (CYP1A2, CYP2C19, CYP2C9, CYP2D6, CYP3A4), or inhibitor of any one or two of the three CYP450s, CYP1A2, CYP2C19, and CYP2C9, except CYP3A4 and CYP2D6 [16,17,18] .

In the present study, 19 compounds were selected for the further study. Molecular weight and lipophilicity score (LogP) of the selected compounds ranged from 62.13 to 331.30 kDa, and 0.08 to 3.69, respectively. The solubility of the compounds was estimated by three methods namely ESOL, Ali, and SILICOS-IT in SwissADME server. The compounds that are soluble atleast in two methods were selected. All the selected phytochemicals were predicted as non inhibitors of all five CYPs. However, three compounds such as cyanidin, malvidin, ellagic acid were identified as CYP1A2 inhibitors, and dibutylphthalate was predicted to inhibit both CYP1A2, and CYP2C19 inhibitor. All the compounds obeyed lipinski rule, veber rule, and egan rule except ellagic acid which obeyed lipinski's rule but did not obey veber's and egan's rule due to topological polar surface area (TPSA) greater than 140 Ų. The compound ellagic acid satisfied other parameters for selection, hence, it was selected for the further study. Physicochemical properties, lipophilicity, water solubility, pharmacokinetics, druglikeness of 16 compounds that were selected for the docking against H. contortus cystatin-1 were listed in Table 2. The toxicity study of the phytochemical compounds revealed that the selected compounds were in the toxicity class 3, 4, 5, and 6. Caffeine, pentanal, and 1-pentanol belonged to toxicity class 3, these compounds are toxic if swallowed, and were not selected for the molecular docking study. The predicted LD 50 and toxicity class of selected phytochemical compounds are given in Table 3. The docking study of 16 compounds showed that cyanidin, epicatechin, caffeic acid, p-coumaric acid, catechin, and malvidin had better binding energy of -4.49, -4.36, -4.18, -4.09, -4.06, and -4.00 kcal/mol, respectively, compared to other ligand molecules. The protein-ligand complexes were generated and analyzed for the stability of ligand within binding pocket. The interaction study of the protein-ligand complex analyzed in PLIP server revealed the multiple hydrogen bonds formed by all the six ligand molecules (Figure 5). The interaction of six Ligands with Haemonchus contortus cystatin-1 was analyzed using BIOVIA Discovery Studio. a) Three-dimensional interaction of the cystatin-1caffeic acid, b) Two-dimensional interaction of the cystatin-1-caffeic acid, c) Three-dimensional interaction of the cystatin-1-catechin, d) Two-dimensional interaction of the cystatin-1-catechin, e) Three-dimensional interaction of the cystatin-1-cyanidin, and f) Two-dimensional interaction of the cystatin-1-cyanidin, g) Three-dimensional interaction of the cystatin-1-epicatechin, h) Two-dimensional interaction of the cystatin-1-epicatechin, i) Three-dimensional interaction of the cystatin-1-malvidin, j) Two-dimensional interaction of the cystatin-1-malvidin, k) Three-dimensional interaction of the cystatin-1-p-coumaric acid, and l) Two-dimensional interaction of the cystatin-1-p-coumaric acid illustrating the key interactions and pocket atoms Catechin followed by epicatechin had the highest number of eight and seven hydrogen bonds indicating stable interaction of Ligands within the binding pocket of the protein. Malvidin had the highest number of six hydrophobic interactions and well as five hydrogen bond interactions and their favorable bond length indicating the strong binding of the ligand. Numbers of hydrogen and hydrophobic bonds formed, their bond length are given in Table 4. Overall, the interaction analysis revealed that all six phytochemicals formed stable complexes with the target protein through synergistic hydrophobic and hydrogen-bond interactions, which are essential determinants of ligand binding affinity and stability of the complex. The orientation of the ligand within the binding pocket is contributed by hydrophobic interactions by reducing the solvent accessibility, and the hydrogen bonds increases the specificity of binding and stability of the ligand within the binding pocket of the protein [19,20] . The hydrogen bond distances observed in the present study (H-A: 1.71-3.62 Å and D-A: 2.59-4.10 Å) are within the acceptable range reported for biologically relevant protein-ligand interactions, indicating favorable binding of the selected compounds. Among the evaluated compounds, catechin, epicatechin, caffeic acid, and malvidin displayed comparatively stronger interaction networks, suggesting greater inhibitory potential than the remaining phytochemicals. These findings support the molecular docking results and provide a structural basis for the potential biological activity of the selected phytochemicals. Nevertheless, molecular docking provides only a static representation of protein-ligand interactions; therefore, molecular dynamics simulations and experimental validation are necessary to confirm the stability, binding behavior, and inhibitory efficacy of these compounds under physiological conditions.

Phytochemicals of C. sinensis (green tea)

Catechin, Epicatechin, Epigallocatechin, Epicatechin gallate, epigallocatechin gallate, Myrecetin-3-rhamnosyl glucoside, Myrecetin-3galactoside, Myrecetin-3-glucoside, Kaempferol-3-galactoside, Kaempferol-3-glucoside, Cyanidin, Delphindin, Pelargonidin, Malvidin,Petunidin, Gallic acid, Chlorogenic acid, Caffeic acid, p-Coumaric acid, Ellagic acid, Quinic acid, Caffeine, Theophylline, Theobromine, Oxalic acid, Tartaric acid, Formic acid, L-malic acid, Ascorbic acid, Lactic acid, Citric acid, succinic acid, Fumaric acid, Dimethyl sulfide, Pentanal, 8-octadecanoic acid methyl ester, Dibutyl phthalate, Hexanal, Acetone, Hexadecane, 1-Pentanol, Heptanal, 2methylbutyraldehyde, 3-myethylbutyraldehyde, Heptadecane, 1-ethylpyrole, 2-methylpropionaldehyde, Naphthalene, Acetic acid

References

1. El-Ashram S, Al Nasr I, Mehmood R, Hu M, He L, Suo X. Haemonchus contortus and ovine host: a retrospective review. Int J Adv Res. 2017;5(3):972-999.
2. Schwarz EM, Korhonen PK, Campbell BE, Young ND, Jex AR, Jabbar A, Hall RS, Mondal A, Howe AC, Pell J, Hofmann A, et al. The genome and developmental transcriptome of the strongylid nematode Haemonchus contortus. Genome Biol. 2013;14(8):R89.
3. Niazi SK, Mariam Z. Computer-aided drug design and drug discovery: a prospective analysis. Pharmaceuticals. 2023;17(1):22.
4. Zhao T, Li C, Wang S, Song X. Green tea (Camellia sinensis): a review of its phytochemistry, pharmacology, and toxicology. Molecules. 2022;27(12):3909.
5. Dwivedi G, Rawal D, Nagda S, Jain T. Anthelmintic activity of tea (Camellia sinensis) extract. Int J Pharm Sci. 2010;1:451-453.
6. Hassan NM, Elsawy BS, Helal MA, Mahmoud MH, ElEzz NM, Shalaby HA. Unveiling of phytochemicals, antioxidants, cytotoxic effects and anthelmintic potency of green tea (Camellia sinensis) beverage against albendazole-resistant Haemonchus contortus. Sci Rep. 2026;16:10486.
7. Waterhouse A, Bertoni M, Bienert S, Studer G, Tauriello G, Gumienny R, Heer FT, de Beer TA, Rempfer C, Bordoli L, Lepore R, et al. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res. 2018;46(W1):W296-W303.
8. Colovos C, Yeates TO. Verification of protein structures: patterns of non-bonded atomic interactions. Protein Sci. 1993;2(9):1511-1519.
9. Ramachandran GN. Stereochemistry of polypeptide chain configurations. J Mol Biol. 1963;7:95-99.
10. Heo L, Park H, Seok C. GalaxyRefine: protein structure refinement driven by side-chain repacking. Nucleic Acids Res. 2013;41(W1):W384-W388.
11. Laskowski RA, MacArthur MW, Moss DS, Thornton JM. PROCHECK: a program to check the stereochemical quality of protein structures. J Appl Crystallogr. 1993;26(2):283-291.
12. Ruwali P, Singh RV, Pandey N. Phytochemical insights into green tea and its therapeutic potential. J Phytopharmacol. 2025;14(5):383-392.
13. Lipinski CA. Rule of five in 2021 and beyond: current status and future directions in drug-like property prediction. Drug Discov Today. 2021;26(3):681-688.
14. Daina A, Michielin O, Zoete V. SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep. 2017;7:42717.
15. Martin YC. A bioavailability score. J Med Chem. 2005;48(9):3164-3170.
16. Banerjee P, Eckert AO, Schrey AK, Preissner R. ProTox-II: a web server for the prediction of toxicity of chemicals. Nucleic Acids Res. 2018;46(W1):W257W263.
17. BIOVIA, Dassault Systèmes. Discovery Studio Visualizer. San Diego (CA): Dassault Systèmes; 2019.
18. Schake P, Bolz SN, Linnemann K, Schroeder M. PLIP 2025: introducing protein-protein interactions to the protein-ligand interaction profiler. Nucleic Acids Res. 2025;53(W1):W463-W465.
19. Klebe G. Protein-ligand interactions as the basis for drug action. In: Drug Design: From Structure and Mode-ofAction to Rational Design Concepts. Berlin, Heidelberg: Springer Berlin Heidelberg; 2025. p. 39-65.
20. Chen D, Oezguen N, Urvil P, Ferguson C, Dann SM, Savidge TC. Regulation of protein-ligand binding affinity by hydrogen bond pairing. Sci Adv. 2016;2(3):e1501240.