Search for cannabinoid receptor 1 antagonists using structure-based virtual screening: identification of natural product hits

Abstract

The diversity of available natural products makes them a rich potential source of new structural classes of cannabinoid receptor 1 (CB1) ligands. CB1 antagonists are clinically established to be effective in treating obesity, obesity-related cardio-metabolic disorders, and substance abuse, but there are currently no marketed CB1 antagonists. We are attempting to discover new chemical classes of CB1 antagonists which may serve as good starting points for drug development. We prepared and validated a three-dimensional (3D) homology model of CB1 (Figure 1) based on a bovine rhodopsin template. The purchasable ZINC subset of 181317 natural products was downloaded from 'www.zinc.docking.org' and prepared using LigPrep that resulted into 298818 compounds. These compounds were then prefiltered to yield 278037 compounds, which were docked using Glide (Schrödinger) into the CB1 model. A total of 192 (Glide score cutoff –9.0) of the top-ranked hits were divided into clusters based on similarity and 18 structurally diverse compounds were selected. Seven of the 18 compounds exhibited > 50% inhibition at 10µM in an in