Objective: To report the preliminary result of virtual screening small molecule inhibitors of tumor necrosis factor-α(TNF-α)based on the traditional Chinese medicine database. Method: Based on the optimized complex structure of TNF-α bound with specific inhibitor (compound 307

PDB code 2AZ5)

computer-aided structure-based virtual screening against Chinese Natural Product Database(CNPD) was conducted to determine the occurrence of herb-based TNF-α inhibitors in the software package Discovery Studio (2.5) of Accerlrys company. The virtual screening results were further filtered by predictive ADME simulation and predictive toxic simulation. Result: According to the dockscore of original inhibitor (compound 307) and the receptor as threshold value

12 drug-like molecules were predicted to have good interactions with TNF-α. Conclusion: It is expected that the TNF-α small molecule inhibitors identified at present study could be helpful for development of new drug to treat rheumatoid arthritis and inflammatory bowel disease.