Bioinformatics and computational exploration of novel inhibitors targeting KRAS G12C mutant protein in lung adenocarcinoma
摘要
The KRAS p.G12C mutation is a major oncogenic driver in lung adenocarcinoma (LUAD), a prevalent subtype of non-small cell lung cancer (NSCLC). The development of sotorasib has provided a new targeted therapeutic option for patients with this mutation.
MethodsKRAS G12C mutations and expression in LUAD were analyzed. A pharmacophore model-based screening of over 1.4 billion conformations was conducted. Molecular docking, drug-likeness, pharmacokinetics, and toxicity predictions were carried out to explore ligands. Binding stability and key residue interactions were evaluated using extra-precision docking, molecular dynamics, H-bond lifetimes, FEL, PCA, and MM-GBSA analyses.
ResultsKRAS mutations were identified in 36% of LUAD samples, with G12C being the most frequent. KRAS p.G12C expression was significantly elevated in mutants (log₂FC = 0.68, p < 0.0001), with strong diagnostic accuracy (AUC = 0.90). A pharmacophore model based on the KRAS p.G12C–sotorasib complex identified 320 compounds. Top candidates, filtered by docking scores, drug-likeness, and toxicity, underwent extra-precision docking and 500 ns molecular dynamics simulations. RMSD, RMSF, Rg, SASA, buriedSASA, hydrogen bond dynamics, FEL and PCA revealed distinct dynamic behaviors among ligands. Key ligands, including PubChem-137,082,465, PubChem-137,304,698, PubChem-137,304,606, and PubChem-154,677,904, demonstrated favorable predicted binding interactions and dynamic stability, with in silico profiles broadly comparable to that of sotorasib. MM-GBSA and per-residue decomposition indicated ARG68, TYR96, and GLN99 as conserved interaction residues critical for ligand stabilization.
ConclusionsThe compounds, including PubChem-137,082,465 and PubChem-154,677,904, exhibited favorable binding and favorable interaction profiles, comparable to sotorasib, positioning them as computationally prioritized candidates for KRAS p.G12C inhibition in LUAD.
Graphical Abstract