4-Hydroxyphenylpyruvate dioxygenase (EC 1.13.11.27, HPPD) is currently one of the popular targets of herbicide research, and herbicides targeting it have shown promising results in treating resistant weeds. However, the long-term application of HPPD inhibitors with in their classical scaffolds inevitably leads to the development of drug resistance. To further delay the underlying development of weed resistance to HPPD inhibitors, we successfully screened HPPD inhibitor pharmacophore with novel chelate structures using computer-aided drug molecular design and obtained 2-trifluoromethyltetrazolium nicotinamide derivatives through active fragment splicing. Among these derivatives, 6-(3,3-diethyl-1-methylureido)-N-(1-methyl-1H-tetrazol-5-yl)-2-(trifluoromethyl)nicotinamide (compound 31), exhibited the most promising in vitro enzyme inhibitory activity, with IC50 value of 0.072 μM, which was approximately five times better than that of the control agent mesotrione (IC50 = 0.363 μM). The crystal structure of the AtHPPD complexed with compound 31 (2.0 Å) and the binding energy calculations of the representative compounds revealed several important interactions of the ligand binding to the target protein, which explained the superior enzyme inhibitory activity of the compounds at the molecular level. In addition, compounds 7 and 31 possessed 100% inhibition against the five target weeds at the tested dosage and both were more effective against Setaria viridis than mesotrione. In general, it is promising to design novel HPPD inhibitors by developing novel chelating structures, and compounds 7 or 31 could be used as the leads for further development of valuable HPPD inhibitors.