Cheminformatics-based identification of phosphorylated RET tyrosine kinase inhibitors for human cancer

Abstract

Background: Rearranged during transfection (RET), an oncogenic protein, is associated with various cancers, including non-small-cell lung cancer (NSCLC), papillary thyroid cancer (PTC), pancreatic cancer, medullary thyroid cancer (MTC), breast cancer, and colorectal cancer. Dysregulation of RET contributes to cancer development, highlighting the importance of identifying lead compounds targeting this protein due to its pivotal role in cancer progression. Therefore, this study aims to discover effective lead compounds targeting RET across different cancer types and evaluate their potential to inhibit cancer progression.

Methods: This study used a range of computational techniques, including Phase database creation, high-throughput virtual screening (HTVS), molecular docking, molecular mechanics with generalized Born surface area (MM-GBSA) solvation, assessment of pharmacokinetic (PK) properties, and molecular dynamics (MD) simulations, to identify potential lead compounds targeting RET.

Results: Initially, a high-throughput virtual screening of the ZINC database identified 2,550 compounds from a pool of 170,269. Subsequent molecular docking studies revealed 10 compounds with promising negative binding scores ranging from -8.458 to -7.791 kcal/mol. MM-GBSA analysis further confirmed the potential of four compounds to exhibit negative binding scores. MD simulations demonstrated the stability of CID 95842900, CID 137030374, CID 124958150, and CID 110126793 with the target receptors.

Conclusion: These findings suggest that these selected four compounds have the potential to inhibit phosphorylated RET (pRET) tyrosine kinase activity and may represent promising candidates for the treatment of various cancers.

Keywords: cancer; docking validation; extra precision docking; high-throughput virtual screening; molecular dynamics simulation; molecular mechanics with generalized Born surface area; phase database; phosphorylated rearranged during transfection tyrosine kinase.

FIGURE 1

Mechanism of action and inhibition of phosphorylated RET tyrosine kinase activity by potential compounds. (A) Role of phosphorylated RET tyrosine kinase in cancer progression. Phosphorylated RET, through mutations and aberrant activation, promotes uncontrolled cell growth and inhibits apoptosis, leading to tumor development and metastasis. RET fusion and mutation are implicated in various cancers, including NSCLC, papillary thyroid cancer (PTC), and medullary thyroid cancer (MTC). (B) Study’s efforts to identify potential compounds that can effectively inhibit phosphorylated RET tyrosine kinase, offering an alternative to the existing inhibitor, vandetanib.

FIGURE 2

Molecular docking scores (kcal/mol) of the top ten hit compounds and native ligand evaluated through different docking approaches. Here, the first column represents the high-throughput virtual screening (HTVS), the second column represents the standard precision (SP), and the third column represents the extra precision (XP) docking scores for respective compounds. The yellow to blue color denotes the elevation of the negative binding score.

FIGURE 3

Docking validation of vandetanib with phosphorylated RET tyrosine kinase (PDB: 2IVU). (A) Native complex form of the crystal structure of phosphorylated RET tyrosine kinase bound to vandetanib. (B) Redocking position of vandetanib with phosphorylated RET tyrosine kinase, demonstrating the docking process. (C) Superimposition of the crystal structure and the docked model, showing the root mean square deviation (RMSD) to illustrate the accuracy of the docking process. (D) Active site pocket and surface view highlighting vandetanib in the active pocket before docking (red) and after docking (green). This panel demonstrates the changes in vandetanib’s position and conformation within the binding site after the docking procedure.

FIGURE 4

Molecular docking interactions between the pRET tyrosine kinase and the four selected compounds, presented in both 3D and 2D formats. (A–D) Interactions of CID 95842900, CID 137030374, CID 124958150, and CID 110126793, respectively.

FIGURE 5

Residual interactions and proximity analysis of the selected four compounds. (A). Various types and numbers of residual interactions observed between the four selected compounds and vandetanib, the control compound. Each bar indicates the count of specific interaction types and shows how the selected compounds differ in their binding characteristics compared to vandetanib. (B). The residues unique to proteins and the selected four compounds are presented alongside the types of contacts established. The closeness of these compounds to proteins identifies the specific residues involved in their interactions.

FIGURE 6

Scatter plots depicting the relationship between ∆ G Bind (binding energy) and various binding energies for selected four compounds. Each panel corresponds to a specific type of binding energy and shows its correlation with ∆ G Bind for the chosen compounds, which are distinguished by different colors. The black line in each panel represents the linear regression fit for the data points. Plot (A): ∆ G Bind vs. ∆ G Bind Coulomb. Plot (B): ∆ G Bind vs. ∆ G Bind Covalent. Plot (C): ∆ G Bind vs. ∆ G Bind H-bond. Plot (D): ∆ G Bind vs. ∆ Gbind Lipo. Plot (E): ∆ G Bind vs. ∆ G Bind Packing. Plot (F): ∆ G Bind vs. ∆ G Bind Solv GB. Panel (G): ∆ G Bind vs. ∆ G Bind vdW. This visualization aids in understanding the relationship between ∆ G Bind and various binding energies for the compounds under study.

FIGURE 7

Overview of pharmacokinetic properties for selected compounds. Herein, (A) provides a comprehensive assessment of various pharmacokinetic properties of the four selected compounds. It covers a wide range of parameters including physicochemical properties, lipophilicity, water solubility, pharmacokinetics, drug-likeness, medicinal chemistry and blood-brain barrier (BBB) permeability. Where, (B) represents the parameters such as, hepatotoxicity, carcinogenicity, immunogenicity, mutagenicity, and cytotoxicity.

FIGURE 8

Graphs representing the MD simulation for the selected protein–ligand complexes, focusing on protein Cα RMSD over a 150 ns simulation period. The compounds CID 95842900, CID 137030374, CID 124958150, and CID 110126793 are depicted in blue, yellow, green, and orange, respectively, in comparison to the control compound vandetanib (black).

FIGURE 9

RMSD values extracted for protein–ligand complex alpha carbon (Cα) atoms of the selected four compounds (ligands) during a 150 ns simulation period. The compounds CID 95842900, CID 137030374, CID 124958150, and CID 110126793 are represented in blue, yellow, green, and orange, respectively, while the control compound vandetanib is depicted in black for comparison.

FIGURE 10

RMSF values of pRET tyrosine kinase were reclaimed from protein Cα atoms of the protein–ligand docked complexes. The compounds CID 95842900, CID 137030374, CID 124958150, and CID 110126793 are represented in blue, yellow, green, and orange, respectively, while the control compound vandetanib is depicted in black for comparison.

FIGURE 11

Number of hydrogen bonds formed of the selected four compounds in a complex with the desired pRET tyrosine kinase and control drug complex during the 150 ns molecular dynamics simulation. The last plot represent the combined hydrogen bond number of selected four compounds CID 95842900, CID 137030374, CID 124958150, CID 110126793, and control compounds of vandetanib, respectively.

FIGURE 12

During the 150 ns MD simulation, the various forms of bonding that took place along the protein–ligand interface are illustrated. The four selected compounds (A) CID 95842900, (B) CID 124958150, (C) CID 137030374, (D) CID 110126793, and control compounds (E) vandetanib are presented.