OBJECTIVE To investigate the structural requirements for effective binding of andrographolide(AGP)and its derivatives(SRJ09and SRJ23)to mutant K-ras for inhibition of exchange factor binding viain silico docking simulations.METHODS The molecular docking studies were carried out by using SiteMap v3.4andGlide v6.6modules(Schrdinger,Inc.).Surface mapping on the 3-D X-ray crystal structures of three mutant K-ras proteins-K-rasG12V(PDB ID:4EPX),K-rasG12C(PDB ID:4LDJ),and K-rasG12D(PDB ID:4DSU),as well as wild-type K-ras protein(PDB ID:4LPK),was performed to generate possible sites for ligand binding.Thirty conformers were generated for each of the studied compounds,and these conformers were docked into each possible binding site in both wild-type and mutant K-ras proteins.The free energy of binding of the compounds with the wild-type and mutant K-ras proteins was performed using prime molecular mechanics with generalized Born and solvent accessibility(MM-GBSA)approach.RESULTS The conformers of AGP,SRJ09 and SRJ23that were found to form the most stable complex inside each possible binding siteas indicated by the highest binding free energy,both in wild-type and mutant proteins,were selected.A common binding site between switchⅠ and Ⅱregions,where a pocket surrounded by amino acid residues Lys5,Leu6,Val7,Ser39,Asp54,Leu56,Tyr71,Thr74,and Gly75,was found in all K-rasG12 mutants.This site corresponds to the hydrophobic binding pockets having aliphatic side-chain portionsas found previously for other Ras binders,which are located betweenα-helix 2 and the core β-sheets(between switchⅠ and Ⅱregions).This common binding pocket was not observed in the wild-type K-ras.A binding pocket adjacent to switchⅡregion(amino acid 60-72),where all ligands bind well,was found instead.All compoundsanchor well inside the common binding pocket in each of the K-fasG12 mutants and these compounds showed the strongest binding interactions to K-fasG12 C.SRJ09 and SRJ23 showed stronger binding interactions to both wild-type and mutant K-ras proteins as compared with the parent compound.Overall,the compounds displayed higher binding energies toall three mutant proteins as compared to their wild-type counterpart.CONCLUSION AGP,SRJ09,and SRJ23 are potential K-ras-targeting anti-cancer agents.The compounds target both wild-type and mutant K-ras but they bind to a different binding pocket in the wild-type protein.Both binding pockets found in wild-type and mutant K-ras involve switchⅡ region that binds the guanine nucleotide exchange factor(GEF)such as Son of Sevenless.These suggest a possible inhibition of exchange factor binding to both wild-type and mutant K-ras proteins.Lower binding energies of the compounds to wild-type K-ras protein suggest a transient binding and inhibition.Stronger binding of all compounds to mutant K-ras proteins could lead to more targeted and prolonged inhibition. OBJECTIVE To investigate the structural requirements for effective binding of andrographolide (AGP) and its derivatives (SRJ09 and SRJ23) to mutant K-ras for inhibition of exchange factor binding via in silico docking simulations. METHODS The molecular docking studies were carried out by using SiteMap v3. Surface mapping on the 3-D X-ray crystal structures of three mutant K-ras proteins-K-rasG12V (PDB ID: 4EPX), K- rasG12C (PDB ID: 4LDJ), and K-rasG12D (PDB ID: 4DSU) were performed to generate possible sites for ligand binding. Thirty conformers were generated for each of the studied compounds, and these conformers were docked into each possible binding site in both wild-type and mutant K-ras proteins. The free energy of binding of the compounds with the wild-type and mutant K-ras proteins was performed using prime molecular mechanics with generalized Born and solvent accessibility (MM-GBSA) approach.RESULTS The conform ers of AGP, SRJ09 and SRJ23that were found to form the most stable complex inside each possible binding siteas indicated by the highest binding free energy, both in wild-type and mutant proteins, were selected. A common binding site between switch I and II regions, where a pocket surrounded by amino acid residues Lys5, Leu6, Val7, Ser39, Asp54, Leu56, Tyr71, Thr74, and Gly75 was found in all K-rasG12 mutants.This site corresponds to the hydrophobic binding pockets has aliphatic side-chain portionsas found previously for other Ras binders, which are located between α-helix 2 and the core β-sheets (between switch Ⅰ and Ⅱ regsions). This common binding pocket was not observed in the wild-type K-ras. A binding pocket adjacent to switch Ⅱ region acid 60-72), where all ligands bind well, was found instead. All compounds are well inside the common binding pocket in each of the K-fasG12 mutants and these compounds showed the strongest binding interactions to K-fasG12 C. SRJ09 and SRJ23 showed stronge r binding interactions to both wild-type and mutant K-ras proteins as compared with the parent compound. Overall, the compounds displayed higher binding energies to all three mutant proteins as compared to their wild-type counterpart. CONCLUSION AGP, SRJ09, and SRJ23 are potential K -ras-targeting anti-cancer agents. these compounds target both wild-type and mutant K-ras but they bind to a different binding pocket in the wild-type protein. Both binding pockets found in wild-type and mutant K-ras involve switch Ⅱ region that binds the guanine nucleotide exchange factor (GEF) such as Son of Seven compounds. These suggest a possible inhibition of exchange factor binding to both wild-type and mutant K-ras proteins. Lower binding energies of the compounds to wild-type K -ras protein suggest a transient binding and inhibition. tronger binding of all compounds to mutant K-ras proteins could lead to more targeted and prolonged inhibition.