Background: PI3K/Akt/mTOR was an important signal transduction pathway, activated
in tumor cells. The dual inhibitors of mTORC1 and mTORC2 could suppress the tumor growth and
angiogenesis by inhibiting the abnormal signal pathway. To deeply understand the structure-activity
relationship of urea-morpholinopyrimidine analogues as inhibitors of mTOR, molecular modeling
studies include Three-Dimensional Quantitative Structure-Activity Relationship (3D-QSAR) and
molecular docking was carried out.
Methods: CoMFA CoMSIA and Topomer CoMFA methods created by 3D-QSAR were used
to build stable model, and significant correlation coefficients for dual inhibitors (q2 = 0.750,
r2 = 0.931 for CoMFA model; q2 = 0.745, r2 = 0.927 for CoMSIA model; q2 = 0.780, r2 = 0.950 for
Topomer CoMFA model) were obtained. At the same time, the generated models were validated using
the test sets (external predictive correlation coefficient rpred
2 = 0.696 for CoMFA; rpred
2 = 0.794 for
Result: These correlation coefficients confirmed that these models were correct and could be used
to predict new compounds. Afterward, molecular docking study was performed to study the molecular
interactions between these analogues and the receptor protein.
Conclusion: In this study, the selected urea-morpholinopyrimidine compounds were docking into
the active site. Some key amino acid residues such as Lys802, Ser806, Lys808 and Val882 were
determined. The satisfactory results from this study may aid in the research and can be used to design
novel dual inhibitors.