Nek6 is a member of the NIMA (never in mitosis, gene A)-related serine/threonine kinase family that plays an important role in the initiation of mitotic cell cycle progression. This work is an attempt to emphasize the structural and functional relationship of Nek6 protein based on homology modeling and binding pocket analysis. The three-dimensional structure of Nek6 was constructed by molecular modeling studies and the best model was further assessed by PROCHECK, ProSA, and ERRAT plot in order to analyze the quality and consistency of generated model. The overall quality of computed model showed 87.4% amino acid residues under the favored region. A 3 ns molecular dynamics simulation confirmed that the structure was reliable and stable. Two lead compounds (Binding database ID: 15666, 18602) were retrieved through structure-based virtual screening and induced fit docking approaches as novel Nek6 inhibitors. Hence, we concluded that the potential compounds may act as new leads for Nek6 inhibitors designing.
Mitotic errors and misregulation of cell cycle process are considered to be an important characteristic of human cancer. The progress of valuable and successful cancer therapies depends mainly on the recognition of physiologic targets that are primarily involved in the regulatory mechanism of cell cycle progression [
In recent years, the members of NIMA-related kinases (Nek) family have contributed to various aspects of mitotic progression and cytokinesis [
In this context, we elucidate the structural information of Nek6, which may be a new drug target for developing inhibitors against cancers by means of a homology modeling approach pursued by a molecular dynamic simulation in order to explore the stability of the protein. However, to date, the structural or drug targeting information against human Nek6 is unavailable. In addition, we calculated the binding site of protein to identify drug-like molecules that acquire enhanced binding energies and pharmacokinetic properties for this Nek6 through high throughput virtual screening. Therefore, the drug-like molecules through screening procedure may act as novel leads for Nek6 inhibitors designing.
Homology modeling is an efficient method for 3D structure prediction and quick experimental design for docking studies. The crystal structure of Nek6 protein is currently unavailable in the Protein Data Bank (PDB). Hence, homology modeling studies have been conducted based on high-resolution crystal structures of homologous proteins. The Nek6 protein sequence was retrieved from Uniprot (Accession no. Q9HC98) and it contains 313 amino acid residues. A sequence similarity search for the protein against other sequences with available structural information was performed using the NCBI BLAST. Crystal structure of APO human Nek7 (PDB ID: 2WQM with 2.10
The constructed Nek6 structure was validated by the inspection of phi/psi distributions of Ramachandran plot obtained through PROCHECK analysis [
The molecular dynamics (MD) simulation of modeled protein was performed using the program Desmond v 3.0 as implemented in Schrödinger package [
The active sites (binding pockets) and functional residues of Nek6 were identified and characterized by Site-Map module from Schrodinger package [
The virtual screening workflow (Maestro, Schrödinger, 2009) [
A mixed molecular docking/dynamics protocol, called induced-fit docking (IFD) was performed by the following steps [
Homology modeling is usually the method of choice when a clear relationship of homology between the sequence of target protein and at least one known structure is found. This approach would give reasonable results based on the assumption that the tertiary structures of two proteins will be similar if their sequences are related [
Sequence alignment of Nek6 (target) with Human APO Nek7 protein (template). Asterisks indicate identical amino acids, dots indicate similar amino acids.
Three-dimensional (3D) structure of Nek6 based on known template protein structure. The structure is shown in secondary structure mode using Pymol.
The backbone conformation of the refined model was validated using Ramachandran plot obtained through PROCHECK. The distribution of the phi and psi angles for the amino acid residues was represented by Ramachandran plot. The generated model was found to be highly plausible; hence, 0.4% (only one residue) of residues alone was found to span the disallowed region of Ramachandran plot (Figure
Percentage residues of Nek6 and Nek7 proteins predicted by Ramachandran plot statistics.
Statistics | Percentage of residues in Nek6 (target) | Percentage of residues in Nek7 (template) |
---|---|---|
Residues in most favored region | 87.4 | 90.8 |
Residues in additionally allowed region | 11.7 | 8.8 |
Resides in generously allowed region | 0.4 | 0.4 |
Residues in disallowed region | 0.4 | 0.0 |
The stereochemical spatial arrangement of amino acid residues in the modelled 3D structure of Nek6 in the favoured region of the Ramachandran plot.
Prosa energy profiles calculated for Nek6 model (a) and the Nek7 crystal structure (b).
Superimposition of modeled Nek6 (target) on Nek7 (template) using the 3d-SS tool. In this wireframe diagram, yellow represents the target and green represents the template.
To gain insight into the stability and dynamic properties of the homolog structures, explicit solvent MD simulation is performed [
Calculated root mean square deviation (RMSD) graph of backbone atoms of Nek6.
The best active site (binding pocket/site) was preferred based on the site score and hydrophobic/hydrophilic areas, which holds better binding cavity. The binding site residues of modeled Nek6 predicted by Sitemap yields 16 amino acid residues in the binding pocket, namely, Ile51, Arg53, Gln55, Lys74, Glu123, Ala125, Asp129, Ser131, Lys135, Lys174, Asn177, Asp190, Gly192, Thr210, Tyr212, and Ser245 (Supplementary Figure S3). These binding site residues may be involved in the binding of substrate and small molecule inhibitors. Thus, all these residues were confirmed as Nek6 active site residues and picked to generate grid in the centroid of these residues for virtual screening approach [
One of the most widely used methods for virtual high throughput screening is docking of small molecules into active site of protein target and subsequent scoring. The Binding database was screened against the putative active site of Nek6 protein using virtual screening work flow (Maestro, Schrödinger, 2009). The ligands were prepared at pH
Extra Precision (XP) results for the ten lead compounds obtained through virtual screening approach.
S. no. | Lead moleculesa | Glide score | Glide energy |
H-bond interactions | Interacting amino acids |
---|---|---|---|---|---|
1 | 15666 | −12.024 | −69.773 | 6 | Ile51, Lys74, Ala125, Lys174, Gly192, Thr250 |
2 | 18602 | −11.095 | −70.845 | 6 | Arg53, Arg53, Lys74, Asp129, Lys174, Asp190 |
3 | 18603 | −10.357 | −65.248 | 6 | Gln55, Lys74, Lys174, Asn177, Asp190, Asp190 |
4 | 18592 | −10.189 | −76.731 | 5 | Arg53, Lys174, Lys174, Tyr212, Ser245 |
5 | 18596 | −10.006 | −72.922 | 8 | Ile51, Ile51, Lys74, Glu123, Ala125, Ser131, Lys135, Lys174 |
6 | 18594 | −9.560 | −76.319 | 6 | Ile51, Lys74, Glu123, Alu125, Lys174, Asn177 |
7 | 18599 | −9.477 | −69.921 | 7 | Arg53, Arg53, Lys74, Ala125, Asp129, Lys175, Asn177 |
8 | 13581 | −9.333 | −56.474 | 4 | Lys74, Lys135, Lys174, Asp190 |
9 | 18597 | −9.253 | −72.195 | 7 | Ile51, Arg53, Lys74, Glu123, Ala125, Lys174, Asn177 |
10 | 18598 | −9.040 | −68.177 | 4 | Lys74, Ala125, Lys174, Asp190 |
Among the ten compounds, the compounds ID 15666 and 18602 have higher binding affinity with high Glide score of −12.024, −11.095 and favorable Glide energy of −69.773, −70.845 kcal/mol, respectively. The compounds id 15666 and 18602 were generally known as INH-NAD Adduct and Mycophenolic Adenine Dinucleotide (MAD) Analogue, 37. The docked protein-ligand complexes with hydrogen bond interactions are shown in Figure
Summary of IFD results for ten best lead molecules.
S. no. | Compound ID | Glide score | Glide energy |
IFD score | Interaction amino acids |
---|---|---|---|---|---|
1 | 15666 | −14.934 | −90.700 | −609.466 | Ile51, Ala125, Lys174, Leu191, Lys74, Thr210, Gln244 |
2 | 18602 | −12.262 | −78.495 | −601.331 | Arg53, Lys174, Asn177, Asp129 |
3 | 18603 | −13.177 | −81.516 | −603.036 | Asp129, Arg53, Lys174, Leu193, Lys74, Ala125 |
4 | 18592 | −14.058 | −89.051 | −605.962 | Gln55, Arg53, Ile51, Lys74, Ala176, Asn177, Ala125 |
5 | 18596 | −13.395 | −81.759 | −606.784 | Pro211, Lys174, Gly192, Ser57 |
6 | 18594 | −13.828 | −83.107 | −606.325 | Ala176, Lys174, Ala125, Arg53, Lys74 |
7 | 18599 | −11.869 | −90.037 | −605.996 | Thr210, Lys174, Asn177, Lys74, Arg53, Ile51, Glu123 |
8 | 13581 | −12.211 | −65.306 | −604.068 | Asp126, Ala127, Asp129, Asp190 |
9 | 18597 | −13.041 | −82.403 | −605.047 | Tyr212, Thr210, Lys174, Lys74, Asp190, Glu123 |
10 | 18598 | −13.709 | −84.140 | −606.432 | Lys174, Leu193, Arg53, Ile51, Ala125 |
The binding mode of compounds (a) 15666 and (b) 18602 with Nek6 obtained from virtual screening approach. Hydrogen bonds are shown in dotted pink lines.
The binding mode of compounds (a) 15666 and (b) 18602 with Nek6 obtained from IFD approach. Hydrogen bonds are shown in dotted black lines.
The comparison of Glide XP results obtained through virtual screening and IFD for the compound (id 15666) clearly shows that they possess less glide score of −12.024 when compared to IFD glide score (−14.934). Similarly, the compound (id 18602) also exhibits low Glide XP score (−11.095) when compared to IFD (−13.395) score, respectively. The glide energy also varies from −69.773 kcal/mol to −90.700 kcal/mol for both compounds. The scores and energies obtained from both studies for all ten compounds are shown in Figure
The graphical comparison of Glide score (a) and Glide energy (b) of ten lead compounds obtained through virtual screening and IFD approach.
In this study, we have developed a homology model of Nek6 based on known crystal structure of Human APO Nek7 protein, as expected from its sequence similarity. After the energy minimization, molecular dynamics simulation was carried out for the modeled structure which results in the most reliable model and this stable structure is further used for receptor ligand interaction analysis. Using Binding database, the high throughput virtual screening was performed which resulted in 10 hit compounds against Nek6 protein. The Glide XP and IFD docking simulations reveal that each compound was bound to the similar region on the active site of Nek6 protein. Hence, we reported that the compounds 15666 and 18602 were good inhibitors against Nek6 protein through its Glide score and Glide energy and will be helpful for the development of drug candidate against Nek6 protein.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors would like to thank the Department of Bioinformatics, Alagappa University, Karaikudi, India, for the support and facility provided for this study.