Identification of novel potential B4GALT6 inhibitors through a large scale virtual screening

Recently, the crucial role of the glycolipid Lactosylceramide (LacCer) has been outlined in the evolution of the progressive form (PMS) of Multiple Sclerosis (MS). LacCer is synthetized by the enzyme B4GALT6, a galactosyl transferase expressed exclusively by reactive astrocytes, and acts in an autocrine manner, driving further inflammation and promoting neurodegeneration. The blockage of B4GALT6 caused a halt to the progression of the disease in established models of MS [1]. Despite these promising results, to date no selective inhibitors acting on this enzyme have been developed. On these grounds, a virtual screening study on B4GALT6 was carried out with the aim of finding novel inhibitors through the screening of an extended dataset of purchasable compounds. Since an experimental structure of the enzyme is not available, two models were generated through a homology modelling approach, mimicking two important conformational states of the enzyme. Then, a docking protocol was developed and optimized, using a purposely collected database including presumed B4GALT6 inhibitors. This protocol was based on the binding space concept and employed an enrichment factor optimization algorithm to create consensus models [2]. These models were then utilized to virtually screen the SPECS screening database, composed of over 207.000 molecules. Based on the computational results, the most promising candidates were further studied to characterize the interaction with the target, assessing the potential inhibition properties