Molecular Docking Study of Selected Antimuscarinic Drugs Against 3D Modeling Candida albicans Rrp9 Protein Using GOLD Revealed a New Antifungal Strategy

3. D Structure modeling

The predicted homology model of the target protein’s accuracy was closely related to the percentage of identity and the optimal alignment of the sequences between the template and target. The results from BLASTP against PDB in search of template structure have revealed the similar structures with high sequence identity. RNA binding Protein of Saccharomyces cerevisiae (PDB ID: 4J0X) had 45% identity and 66% similarity with the sequence of our target protein Rrp9 from C. albicans (Q5A213_CANAL). On the basis of sequence alignment of the selected template RNA binding Protein of Saccharomyces cerevisiae (PDB ID: 4J0X) and target sequence Rrp9 from C. albicans, the modeling of the protein was carried out to generate high quality models. In the inbuilt programs of DS perform sequence alignment and homology modeling, the sequence alignment is done using sequence analysis protocols called “Align multiple sequences”. The protocol inputs are target sequence and 4J0X template, which aligns them and gave the alignment output as shown in Figure 3. The initial models of the Rrp9 was built using protein modeling protocol called “build homology model” which uses modeler to build homology models. The modeler options are kept default while running ended up with a loop refinement by default modeling process. Of all the developed three homology models of Rrp9 the model with least internal scoring functions, of DS was selected. The build models were B99990001, B99990002 and B99990003. Out of the three models, Model of B99990001 as shown in Figure.4 had the lowest value in PDF (Probability Density Functions) Total Energy (6141.58), PDF Physical Energy (-556.69) and DOPE (Discrete Optimized Protein Energy) score (-40198.65) , which indicated it as the best model. The energy refinement method gave the best conformation to the model. For optimization, the CHARMm forcefield and steepest descent algorithm were applied with 0.001 minimizing RMS gradient and 1000 minimizing steps. Following the above steps of minimization, the protein was minimized using conjugate gradient algorithm preceded by smart minimizer algorithm until the convergence gradient of 0.001 kcal mol-1 is satisfied. The finally refined protein generated was further taken for model validation by PROCHECK, Verify 3D, ProSA and RMSD.

Figure 3. Alignment between the RNA binding Protein of Saccharomyces cerevisiae (PDB ID: 4J0X) and query sequence Rrp9 from C. albicans. Thick blue color represents the conserved regions and light blue color represents variable regions of template and query.

Figure 3. Alignment between the RNA binding Protein of Saccharomyces cerevisiae (PDB ID: 4J0X) and query sequence Rrp9 from C. albicans. Thick blue color represents the conserved regions and light blue color represents variable regions of template and query.

Figure 4. Representation of Modeled Rrp9 structure in N-C terminal representation contains number of helices 4, Strands 47, turns 54.

Figure 4. Representation of Modeled Rrp9 structure in N-C terminal representation contains number of helices 4, Strands 47, turns 54.

Model Validation

The stereo chemical quality and accuracy of the final refined modeled Rrp9 protein was evaluated by Ramachandran plot calculations computed with the PROCHECK, described a good quality model which had over 90% residues in most favored region. The results showed that 93.1% residues were in favored region (Figure 5). Of the total sequence length, 549 amino acid residues in favorable and additionally allowed and only five residues were in generously allowed regions which indicated a good quality model.

In Verify 3D, the certainty of the 3D model was determined in contrast with its own sequence adopting a 3D profile estimated using atomic coordinates of its structure. 3D profiles of equitable protein structure possessing high S scores in contrast with poor structures with low scores. The S score/3D-1D profile score is the sum over all residue positions of 3D-1D score for amino acid sequence of the protein. The validation results of Verify 3D for this model demonstrated that at least 86.51% of the amino acids have scored > = 0.2 in the 3D/1D profile, which is also satisfactory.

Figure 6. The Verify 3D score profile of the homology model (target) protein.

Figure 6. The Verify 3D score profile of the homology model (target) protein.

The overall protein quality was calculated using ProSA based on the scores of all experimentally determined protein chains found in PDB. The program was gave z-score and a plot of the residue energies. The Z-Score plot showed spot of Z score values of proteins determined by NMR (represented in dark blue color) and by X ray (represented in light blue color). The Z-scores of the obtained modeled Structure of Candida Rrp9 was –6.82 (shown in Figure 7). The black dot represent Z-scores of our model that was within the range of scores of similar proteins determined by X ray crystallography derived structures. .

Finally, this was also confirmed by superposition of optimized homology model against the template and the RMSD is estimated using SPDBV. The structural super imposition of the Rrp9 from C.albicans and the template RNA binding Protein of Saccharomyces cerevisiae (PDB ID: 4J0X) is shown in Figure 8. The weighted root mean square deviation was 0.75 Å. This final refined model was used for the identification of active site and for docking of the ligands with the homology model of Rrp9.

Identification of Active Site

The active site residues of the modeled Rrp9 were identified by using DS which employs cavity detection method using eraser algorithm. The predicted active site pocket 1 shown in figure 9 A had a average volume of 500 and had coordinates as X-1.648, Y-14.419, Z—28.890.

Molecular Docking

The binding mode for the taken ligands into the predicted binding site of the homology model Rrp9 was generated by molecular docking using GOLD software. GOLD Fitness score was used to estimate the ligand binding energies. The orientation of protein–ligand complex with the highest GOLD score was chosen for further analysis. Apart from these, hydrogen bond formations are also considered to evaluate the binding efficacy of the ligands with the protein. The dock scores computed by the scoring algorithm are given in Table 3. All the docked compounds showed fitness scores with a range of 18.75 to 62.17 (Table 3). Almost all the compounds showed hydrogen bonding with the binding site residues of the target protein Rrp9, but Promethazine Hydrochloride doesn't form any hydrogen bonds. Even though the compound Promethazine Hydrochloride had good dock scores it does not show any hydrogen bond interactions. The compounds, Dicycloamine Hydrochloride and Solifenacin Succinate showed single hydrogen bond and the compound Hydroxyine hydrochloride had two hydrogen bond formations with the target protein Rrp9. Considering the dock scores and hydrogen bond interactions, the Solifenacin succinate have chosen as the best compound and Hydroxyzine hydrochloride next to that among all the compounds. Though Hydroxyzine hydrochloride formed two hydrogen bonds, it showed least binding affinity with a fitness score of 54.55 compared to that of the Solifenacin succinate having a fitness score of 62.17. The hydrogen bonding interactions and the interacting amino acids along with the hydrogen bond distances were depicted in the Table 4. Figure 9B-9H showed the binding poses predicted for the compounds against Rrp9 protein model.

Figure 9B presented the binding pose of compound Solifenacin succinate against Rrp9, indicating that the compound interacts with Rrp9 by forming hydrogen bond with Arginine275. The N-atom of Arginine275 interacted with the oxygen atom of the compound Solifenacin Succinate(Arg275:NE----O2: Solifenacin Succinate) with a hydrogen bond distance of 2.914 Å . the Solifenacin Succinate compound also showed additional interactions with residues Phe273, Asn173 and Pro234 in binding pocket. They are formed with the oxygen atom of phenyl alanine 273 and eight carbon atom of the compound (Phe273: O---C8: Solifenacin Succinate), between the alpha carbon atom of amino acid proline234 and 24^th carbon atom of the compound (Pro234:CA---C24: Solifenacin Succinate), between the oxygen atom of amino acid aspargine173 and 17^th carbon atom of compound (Asn173:O---C17: Solifenacin Succinate) with a distances of 2.503 Å, 2.356 Å and 2.702 Å respectively.

From the figure 9C, it was revealed that two hydrogen bonds are formed between the amino acid Isoleucine271 and isoleucine 230 and the Hydroxyzine Hydrochloride compound with a hydrogen bond distances of 2.581 Å and 2.597 Å respectively. One hydrogen bond was formed between the oxygen atom of isoleucine 271 and hydroxy group of the compound (Ile271:O---OH: Hydroxyzine Hydrochloride ) and the other between the oxygen atom of isoleucine 230 and hydroxyl group of the compound (Ile230:O---OH: Hydroxyzine Hydrochloride ). The other interactions were formed with the oxygen atom of serine233 and the carbon atom C21 of the compound (Ser233:O—C21: Hydroxyzine Hydrochloride) and proline 234 interacted with C17 of the compounds (Pro234:C---C17: Hydroxyzine Hydrochloride) with a distances of 2.468 Å and 2.752 Å respectively.

From the Figure 9D it was revealed that the hydrogen bond was formed between nitrogen atom of Tyrosine 174 and the oxygen atom of the compound Dicycloamine (Tyr174:N---O2:Dicycloamine) with a hydrogen bond distance of 2.340 Å. Some close contacts were also formed between the protein and the compound. They were formed with the carbon atom of Tyrosine174 and carbon atom number 8 of the compound (TYR174: CB---C8: Dicycloamine), between the carbon atom of amino acid aspargine173 and oxygen atom of the compound (Asn173: C...O2: Dicycloamine) and between the carbon atom of amino acid glutamine509 and 21^st carbon atom of compound (Glu509: CG...C21: Dicycloamine) with a bond distances of 2.603 Å, 2.642 Å and 2.881 Å respectively.

Figure 9E and 9F showed the interactions of the protein Rrp9 with the compounds, Promethazine and Flavoxate indicating some close interactions were formed between the protein and the compounds. From the figure 9E, it was revealed that the 13^th carbon atom of the compound, Promethazine interacted with the carbon atom of tyrosine174 (Tyr174: CB---C13: Promethazine) and nitrogen atom of proline234 (Pro:234: N....C13 : Promethazine) with a distances of 2.750Å and 2.897Å respectively. Figure 9F showed that the oxygen atom of the residue Alanine 232 interacted with 21^st carbon atom and oxygen atom of the compound, Flavoxate (Ala232: O---C21: Flavoxate and Ala232;O---O4: Flavoxate) with a bond distances of 2.621Å and 2.596 Å respectively. And also the oxygen atom and 11^th carbon atom of Flavoxate interacted with the oxygen atoms of phenylalanine273 and Alanine315 (Phe273: O---O3: Flavoxate Hydrochloride and Ala315: O---C11: Flavoxate) with a bond distances of 2.709Å and 2.559 Å respectively.

Figure 9: A. Visualization of active site pocket 1 and Binding poses predicted by GOLD against the homology model Rrp9 with B. Solifencin C. Dicyclomine D. Hydroxygene E. Promethazine F. Flavoxate. Hydrogen bonds are depicted by a green dotted line.