ABSTRACT
BACKGROUND: The tumor pyruvate kinase M2 (PKM2) is involved in the glycolytic pathway of lung cancer and targeting this kinase has been observed to radiosensitize non-small cell lung cancer (NSCLC). OBJECTIVE: An integration of in silico virtual screening and in vitro kinase assay was described to discover novel PKM2 inhibitors from a candidate library containing >400,000 commercially available compounds. METHOD: The method is a stepwise screening scheme that first used empirical strategies to fast exclude those undruggable compounds in the library and then employed molecular docking and molecular dynamics (MD)-based rescoring to identify few potential hits. Subsequently, the computational findings were substantiated using a standard kinase assay protocol. RESULTS: Four compounds, i.e. nalidixic acid, indoprofen, hematoxylin and polydatin, were identified to inhibit PKM2 kinase at micromolar level, with IC50 values of 53, 21, 340 and 128 .M, respectively. CONCLUSION: Structural analysis revealed that hydrogen bonds, salt bridges, π-π stacking and hydrophobic forces co-confer high stability and strong specificity to PKM2-inhibitor binding.
