3-Hydroxy-3-methylglutaryl coenzyme A reductase (HMGR) is a primary target in the current clinical treatment of hypercholesterolemia with specific inhibitors of "statin" family. Statins are excellent inhibitors of the class I (human) enzyme but relatively poor inhibitors of the class II enzyme, which are well-known as a potential target to discover drugs fighting against the invasive diseases originated from S. pneumoniae . However, no significantly effective inhibitors of class II HMGR have been reported so far. In the present study, the reasonable three-dimensional (3D) structure of class II HMGR from S. pneumoniae (SP-HMGR-II) was built by Swissmodel. On the basis of the modeling 3D structure in "close" flap domain form, several novel potential hit compounds out of SPECs database were picked out by using structure-based screening strategy. Especially the compounds 4, 3, and 11 exhibit highly inhibitory activities, with IC50 values of 11.5, 18.5, and 18.1 μM, respectively. Furthermore, the hit compounds were chosen as probe molecules, and their probable interactions with the corresponding individual residues have been examined by jointly using the molecular docking, site-directed mutagenesis, enzymatic assays, and fluorescence spectra, to provide an insight into a new special binding-model located between the HMG-CoA and NADPH pockets. The good agreement between theoretical and experimental results indicate that the modeling strategies and screening processes in the present study are very likely to be a promising way to search novel lead compounds with both structural diversity and high inhibitory activity against SP-HMGR-II in the future.