INTRODUCTION

Venoms are essential in the defense of venomous animals. Venom components activate cells, receptors, and signalling pathways, resulting in discomfort and inflammation [1]. Envenomation by snakes has a huge health and economic impact around the world. Snakebite is a deadly environmental and occupational disease, especially in rural areas of tropical countries, where around 40,000 to 50,000 snakebites are reported each year [2]. The poisonous viper Echis coloratus is indigenous to the Gulf Region and Egypt. It reaches a maximum total length of 75 cm. From sea level to 2,500 meters (8,200 ft) above sea level, it can be found in stony deserts [3]. Sand deserts are devoid of it.

Notable envenoming snakes include Echiscoloratus and Naja n. Nigricollis, whose bites cause both systemic and local pathology [4]. However, when it comes to their pathophysiological activities, the venoms of the two species react differently [5]. The bite of an Echis coloratus (Carpet viper) causes inflammation (swelling, blistering, and necrosis) as well as haemorrhages [6]. Snake venom is a complex mixture of proteins and peptides that serve a number of biological purposes. Regularly used to assess snake envenomation are horses and sheep. Snake venom metalloproteinase (SVMP) is a zinc metalloproteinase that belongs to the extracellular matrix (ECM) protein family [7]. SVMP has been implicated as a mediator for edoema, local tissue injury, inflammation, and bleeding in recent research [8]. Because of the limited haemorrhage and damage to endothelial cells, SVMP is an enzymatic toxin [9].

As a result of the hemorrhaging brought on by SVMP, tissue necrosis, shock, hypotension, hypovolemia, inflammation, and a diminished ability for muscle tissue to regenerate were all caused. The toxin SVMP is considered to be the most significant in the pathology brought on by snake venom. The majority of the time, SVMP disrupts the junction between the basement membrane and the endothelium, which directly affects capillary blood vessels and damages them, causing hemorrhage [10, 11]. It is obvious that SVMP-targeted medication therapy would have a positive consequence by lowering patient mortality rates [12].

Antivenin is the only specific treatment for snakebite envenoming. Investigations have shown that antivenin is ineffective against venom-induced local tissue damage and frequently causes clinical symptoms like anaphylactic shock. As a result, alternative, synthetic or natural snake venom inhibitors are required to help get around antivenin limitations. The overall local tissue damage after envenomation may be significantly reduced by inhibition of SVMPs and other components. Finding SVMP inhibitors is now a top research priority as a result. Studies have revealed that snake venom enzymes are inhibited by plant secondary metabolites.

The interaction between the receptor protein SVMP and the ligand is demonstrated using a molecular docking approach. Ligands are small molecules that bind to the active pockets of protein. These pockets are the specific sites where the ligand can interact with the receptor protein (SVMP). There are numerous unique pocket active sites on which binding can occur in receptor atoms. There are different binding pockets, and binding modes. Docking also provides information on the strength of the binding, the ligand-receptor complex energy, and the complex docking score calculation [13]. The most advantageous aspect of docking is its use in medicine. The algorithm and energy scoring function are used to anticipate different possess when ligand binds with the receptor protein.

The search for investigation of inhibitors is a crucial phase in medication development. By discovering inhibitors against the target protein (SVMP), this research aids in evaluating the inhibitors blocking the effect of SVMP and are useful for drug designing to overcome antivenin restrictions. Furthermore, the current study predicts the inhibitors that showed strong binding affinity during docking and good stability with receptor protein in the process of molecular dynamic simulation.

MATERIALS AND METHODS

Protein Preparation

Amino acid sequence of SVMP was retrieved from Uniprot [14].  Using computer algorithms, the amino acid sequence can be used to create a three-dimensional model of a protein. I-TASSER was used to model the 3D structure of SVMP [15]. The above server uses multiple-threading alignments and iterative prototype fragment assembly simulations to create 3D protein structures. I-TASSER provides confidence scores to determine the accuracy of the model. The predicted tertiary structure was checked for errors using the ProSA-web, verify 3D, ERRAT server, and Ramachandran plot analysis [16-19]. The average quality score of the structure is described by ProSA-web. Furthermore, non-bonded atom-atom interactions were evaluated using the ERRAT server. The energetically permitted and prohibited dihedral angles of amino acid residues, phi (φ) and psi (ψ), were evaluated using the Ramachandran plot. Protein structure preparation for docking was carried out using the Molecular Operating Environment (MOE) [20].

Ligand Database Preparation

A thorough literature search was done to find phytochemicals that have been noted to be effective against snake bites. Chemical structures of phytochemicals were obtained from the MPD3 database, the Pubchem database, the MAPS database, and the Zinc database in a variety of ligand file formats, including mol, sdf, and mol2 [21-23]. The Protonate3D module was used in MOE to modify these ligand structures by adding partial charges. The MMFF94X force field was used to lower the energy of each ligand. The ligands were then added one by one to the MOE ligand database for docking.

Active Sites Prediction

CPORT, a tool included in the Haddock interphase, was used to predict active sites of the target protein (SVMP) [24]. These active site pockets will bind to the target protein, at which point the ligand will bind to inhibit the protein's activity. Donor, acceptor, hydrophobic, hydrophilic, and metal-binding domains make up the sites. For molecular docking, the most precise predicted binding pocket was chosen.

Molecular Docking

A ligand database of over 2500 phytochemicals was docked using the MOE Dock tool at the SVMP protein's designated docking sites. The triangular matcher algorithm was used to find 1,000 best poses for docked molecules using the default ligand placement method. The London DG scoring function was used to redo the scoring of simulated poses [25]. The London dG algorithm, which determines final binding energy using the Generalized Born Solvation Model while maintaining rigid receptor residues, was used to further reduce the top ten scored poses per molecule. In addition, the compounds' binding affinity, S-score, and Root-Mean-Square Deviation (RMSD) were used to rank them. Only substances with a high docking score that bind to the SVMP protein's active residues were to be chosen from the top-ranked poses.

Ligand-Protein Interaction Analysis

To better comprehend the most intricately docked complex protein-ligand interactions, 2D plots of ligand-protein interactions were examined using the MOE LigX tool [26]. In the SVMP active site, it generates a two-dimensional graph of hydrogen bonds, electrostatic interactions, hydrophobic interactions, and all of these factors contribute to the increased affinity of drug-like molecules. 3D images of SVMP protein inhibitor complexes were created using UCSF Chimera [27].

Drug scan/ADME Toxicity

Using the drug scanning tool on the Molinspiration server, a computational approximation of the drug probability of docked phytochemicals was discovered within the parameters established by Lipinski's Rule [28, 29]. The ADMETSAR server was used to simulate the absorption, deposition, metabolism, excretion, and toxicity profiles of these hits  [30]. Additionally evaluated were AMES toxicity and the inhibitors' potential carcinogenicity.

MMGBSA/MMPBSA Analysis

Using the Prime module of the Schrodinger suite and the OPLS-2005 force field, it was possible to calculate the binding free energies of protein-ligand complexes [31, 32]. Binding free energies were assessed using the MM-GBSA method and the MM-PBSA.py program, with 250 snapshots taken throughout the final 20 ns of simulation at regular intervals. The complex free energy was subtracted from the total of the ligand and protein free energies to determine the binding free energy.

Where ΔG(binding) denotes the binding free energy and ΔG(complex), ΔG(ligand), and ΔG(protein) denote the complex, ligand, and protein free energies, respectively.

MD Simulations

MD simulations were conducted using the Desmond software [33]. The OPLS-2005 force field and the TIP4P water model were used to analyze the protein-ligand interactions in this system [34]. A 10 Å buffer region was made between the orthorhombic water box's sides and the protein atoms using this method [35]. The systems were defused by adding Na+ ions after the overlapping water molecules were removed. The OPLS-2005 force field was used to calculate energy. A 2.0 fs value was obtained by maintaining the temperature at 300 K throughout the integration step. The consistency of the SVMP protein in its natural motion was tracked using the root mean square fluctuations (RMSF), root mean square deviation (RMSD), radius of gyration (RoG), and solvent accessible surface area (SASA). After saving the synchronized file at intervals of 5000 ps for up to 100 ns, the results were examined using the Nagasundaram et al. method [36].

RESULTS AND DISCUSSION

Protein Preparation

The SVMP protein's amino acid sequence was obtained from Uniprot (UNIPROT ID: E9JGE0) in fasta format. The I-TASSER server was employed to predict SVMP's 3D structure. I-TASSER was used to develop the top five models, from which the best model was chosen, with a C-score of around -5.04, indicating high quality. The model's quality was confirmed by Ramachandran plot analysis, which revealed that 90.5 % and 7.1 % of overall amino acids are found in the most favorable and allowed regions, respectively, while only 0.4 percent are found in the disallowed region. Further analysis revealed that the model was of good quality, with a Z-score of -6.86, a compatibility score of 81.62 % (as determined by Verify 3D), and a quality factor of 80.8 (as determined by ERRAT).

The structure was prepared for docking using MOE. MOE was used to perform 3D protonation, water molecule removal, and energy minimization.

Active Site Prediction

Cport was used to select binding pockets against the target protein. The reported active residues for the receptor SVMP were discovered to be Pro 1, Tyr3, Tyr 40, Arg 41, Asn 44, His 46, Glu 209, Leu 210, Leu 211, Gln 127, Asp 220, Pro 221, Thr 223, Ser 232.

Molecular Docking

With the aid of a database of phytochemical ligands, the SVMP protein was docked. An S-score function was used to calculate and represent the hydrogen bonding strength, maximum binding pocket occupancy, lowest Gibbs free energy, and potential non-covalent interactions in order to rank the docked compounds. Out of 25, 000 docked molecules, the best docking molecules were chosen. According to threshold-based criteria, the ligand had to bind to the target SVMP protein using all of the hotspot conserved residues in the binding pocket and achieve the necessary S-score values (the lower the score, the greater the interactions and affinities).

The active sites of the SVMP were found to bind Rutamarin, Enterodiol, Butyl butyrate, Colchicine, Sanggenon A, Quercetin, Campesterol, and Cholesterol with high affinity (Figures 3 and 4). The minimum binding energy of these selected phytochemicals ranged from -19.87 kcal/mol to -13.85 kcal/mol. The scoring function and minimum binding energy of each docked ligand are shown in Table 1.

Rutamarin, on the other hand, was ranked first among the revealed SVMP binding residues due to its high affinity and binding score. Most phytochemicals seem to have a strong affinity for the amino acids Arg 41 and Thr 223, implying that these are the most active residues.

Drug Scan/ADMET Results

The Lipinski Rules of Five-based Molinspiration server was used to predict the drug-likeness of screened compounds. The chosen candidates exhibited drug-like properties and did not violate the "rule of five” (Table 2).

The pharmacokinetic properties of selected compounds were analyzed using the admetSAR server to further verify the potential of drug likeliness, and the results are presented in Table 3.