A Random Forest Model for Predicting Allosteric and Functional Sites on Proteins.

We created a computational method to identify allosteric sites using a machine learning method trained and tested on protein structures containing bound ligand molecules. The Random Forest machine learning approach was adopted to build our three-way predictive model. Based on descriptors collated for each ligand and binding site, the classification model allows us to assign protein cavities as allosteric, regular or orthosteric, and hence to identify allosteric sites. 43 structural descriptors per complex were derived and were used to characterize individual protein-ligand binding sites belonging to the three classes, allosteric, regular and orthosteric. We carried out a separate validation on a further unseen set of protein structures containing the ligand 2-(N-cyclohexylamino) ethane sulfonic acid (CHES).

Cambinol, a novel inhibitor of neutral sphingomyelinase 2 shows neuroprotective properties.

Ceramide is a bioactive lipid that plays an important role in stress responses leading to apoptosis, cell growth arrest and differentiation. Ceramide production is due in part to sphingomyelin hydrolysis by sphingomyelinases. In brain, neutral sphingomyelinase 2 (nSMase2) is expressed in neurons and increases in its activity and expression have been associated with pro-inflammatory conditions observed in Alzheimer's disease, multiple sclerosis and human immunodeficiency virus (HIV-1) patients. Increased nSMase2 activity translates into higher ceramide levels and neuronal cell death, which can be prevented by chemical or genetic inhibition of nSMase2 activity or expression. However, to date, there are no soluble, specific and potent small molecule inhibitor tool compounds for in vivo studies or as a starting point for medicinal chemistry optimization. Moreover, the majority of the known inhibitors were identified using bacterial, bovine or rat nSMase2. In an attempt to identify new inhibitor scaffolds, two activity assays were optimized as screening platform using the recombinant human enzyme. First, active hits were identified using a fluorescence-based high throughput compatible assay. Then, hits were confirmed using a 14C sphingomyelin-based direct activity assay. Pharmacologically active compounds and approved drugs were screened using this strategy which led to the identification of cambinol as a novel uncompetitive nSMase2 inhibitor (Ki = 7 µM). The inhibitory activity of cambinol for nSMase2 was approximately 10-fold more potent than for its previously known target, silence information regulator 1 and 2 (SIRT1/2). Cambinol decreased tumor necrosis factor-α or interleukin-1 ß-induced increases of ceramide and cell death in primary neurons. A preliminary study of cambinol structure and activity allowed the identification of the main structural features required for nSMase2 inhibition. Cambinol and its analogs may be useful as nSMase2 inhibitor tool compounds to prevent ceramide-dependent neurodegeneration.