Designing of Protein Structural of Ranpirnase Based on Bovine Pancreatic Ribonuclease with Using Molecular Dynamic and Static Simulation

Animal production is not only restricted to food products but some therapeutics such as bovine pancreatic ribonuclease (RNase A) (15). The RNase A is known as a powerful enzyme in ribonuclease family that is used in biotechnology industry (1). It has a potential to be used as an immunotoxin although there are two main reasons that RNase A is not suitable for immunotoxin engineering. The lack of ability to evade ribonuclease inhibitor (RI) and very weak cell penetration, which are essential properties for an immunotoxin (1). One of the small member of ribonuclease family is ranpirnase, originated from the Northern Leopard Frog (Rana pipiens) that has suitable charateristics for immunotoxin engineering (24, 29). It was showed that Ranpirnase cytotoxicity was five times less than RNase A (26). Therefore, detection of enzyme properties of RNase A and Ranpirnase could be useful for engineering efficient immunotoxins from Ranpirnase. The aim of this study was in silico engineering of Ranpirnase enzyme based on properties of RNase A to design an efficient immunotoxin with the high cell penetration, low cytotoxicity, ability to evade RI and structural stability.

 All protein structures required for performing this study were extracted from Protein Data Bank (PDB) (http://www.rcsb.org). The PDB files related to ribonuclease inhibitor (RI) with accession number of (10.2210/pdb2BNH/pdb), the PDB file related to ribonuclease inhibitor bonded to RNase A enzyme with accession number of (10.2210/pdb1DFJ/pdb), the PDB file related to Ranpirnase enzyme (pdb1YV6/pdb/10.2210) and the PDB file of RNase A enzyme with accession number of (10.2210/pdb2K11/pdb) were selected. PyMOL Software and ClusPro online server were used for docking mentioned enzymes with ribonuclease inhibitor. In PyMOL Software (ver 1.8x), RNase enzyme of PDB file related to ribonuclease inhibitor connected to RNase A enzyme (10.2210/pdb2BNH/pdb) was manually replaced by ranpirnase enzyme. Also, docking prediction related to each enzyme with enzymatic inhibitor was performed by introducing ligand and receptor to ClusPro Software. Then, obtained results were used in molecular dynamic (MD) studies. All stages of MD simulation were performed using GROMACS Software (version 5) in Linux 17.2 environment and CHARMM force filed (27). In summary, protein structures of this study were placed in a cube box filling with more than 6700 water molecules. Ionization was performed to achieve the natural pH of the environment. Extra charge of the system was adjusted by adding appropriate number of ions to the distance of 7 angstrom of protein surface. Minimizing system energy was performed at 300 K for 20 ps. The length of all bonds was limited by Links. Newton's equation of motion matched with a 2fs time interval and atomic characteristics for every 0.5 ps were stored to be analyzed. Dielectric stability is considered to be 1. Simulated temperature is 300 K. A 4-ps simulation at 300 K was performed to investigate dynamic condition of ranpirnase and ribonuclease inhibitor, pancreatic RNase with ribonuclease inhibitor. The structural stability of the simulation was assessed using several geometric parameters per unit time such as: Root mean square deviations (RMSD), root mean square fluctuations (RMSF), and gyration radius. Also, protein structures at different time points were analyzed using PyMOL and VMD computer programs.

 Since the topological structures of ranpirnase enzyme and RNase A are similar, it was expected that binding of RI with ranpirnase enzyme is similar to RNase A enzyme. The docking results showed that RNase A enzyme was bonded to RI through 19 amino acids. This binding with RI is through 5 amino acids for ranpirnase enzyme. The glutamine 11 and serine 89 were most important residue that bonded to RI in RNase A. We found that pyroglutamine 1 and serine 72 are the homolog residue in ranpirnase. The ranpirnase ability to evade RI was obtained with mutation S72A and cytotoxicity and cell penetrate were achieved by K45R, K49R, L55R mutations. The molecular dynamic simulation confirmed the stability of mutant ranpirnase by RMSD, RMSF and Vg analysis. Also, protein charge of surface in mutant ranpirnase was increased in compare to the native ranpirnase.

It was reported previously that ranpirnase enzyme is able to evade RI but our modeling results demonstrated that RI could bind with ranpirnase. The ranpirnase with 4 mutations (K45R, K49R, L55R and S72A) seemed to be more efficient as a suitable toxin and had favorable properties such as evading from RI, cell penetrate, cytotoxicity and protein stability in comparison of native enzyme. Also, we optimized a new approach for designing and engineering of immunotoxins.