Virtual screening of novel reversible inhibitors for marine alkaline protease MP.

Marine alkaline protease (MP,(2) accession no. ACY25898) is produced by a marine bacterium strain isolated from Yellow Sea sediment in China. Previous research has shown that this protease is a cold-adapted enzyme with antioxidant activity that could be used as a detergent additive. Owing to its instability in the liquid state, MP's application in liquid detergents was limited. Therefore, the discovery of reversible MP inhibitors to stabilize the protease was imperative. Here, we used the X-ray structure of MP and recompiled AutoDock 4.2 with refined Zn(2+) characters to screen the free chemical database ZINC. After completing the docking procedure, we applied strategies including the "initial filter", consensus scoring and pharmocophore model to accelerate the process and improve the virtual screening success rate. The "initial filter" was built based on the docking results of boronic acid derivatives validated as reversible inhibitors of MP by our previous studies. Finally, ten compounds were purchased or synthetized to test their binding affinity for MP. Three of the compounds could reversibly inhibit MP with apparent Ki values of 0.8-1.2 mmol. These active compounds and their binding modes provide useful information for understanding the molecular mechanism of reversible MP inhibition. The results may also serve as the foundation for further screening and design of reversible MP inhibitors.

Simultaneous determination of nucleosides and their bases in Cordyceps sinensis and its substitutes by matrix solid-phase dispersion extraction and HPLC.

Nine nucleosides and nucleobases, including uracil, adenine, thymine, uridine, adenosine, thymidine, cytidine, guanosine, and cordycepin in natural Cordyceps sinensis, cultured Cordyceps mycelia, and Cordyceps fruiting bodies were extracted by matrix solid-phase dispersion (MSPD) and determined by HPLC. The experimental conditions for the MSPD extraction were optimized. Florisil was used as dispersant, petroleum ether as washing solvent, and methanol as elution solvent. The Florisil-to-sample ratio was selected to be 4:1 and no additional clean-up sorbent was needed. The calibration curves had good linear relationships (r > 0.9997). The LOD and LOQ were in the range of 12～79 and 41～265 ng/mL, respectively. The intra- and interday precision were lower than 8.3%. The recoveries were between 61.5 and 93.2%. The present method consumed less sample compared with ultrasonic extraction and heating reflux extraction (HRE). The extraction yields obtained by using the present method are much higher than those obtained by UE and comparable to those obtained by HRE.