s In Silico Target Identification and Molecular Docking Studies of Natural Cytotoxic Compound Borivilianoside H

Background: Borivilianoside H is a naturally occurring anti-cancer compound with known cytotoxicity against human colorectal cancer cell line (HCT-116) and human adenocarcinoma cell line (HT-29). The present study describes the pharmacophore modelling, molecular docking, and molecular dynamics simulation approaches to predict the target proteins of borivilianoside H along with its binding affinity to the selected proteins. Methods: A 3-dimensional structure of borivilianoside H was constructed using Avogadro from its 2-D coordinates retrieved from the Pubchem Compound database. Target proteins associated with cancers were identified based on the 95% normalized fit score of PharmMapper. The crystal structures of the targets were retrieved from Protein Data Bank and molecular docking was performed with Autodock Vina 1.1.2. MD simulations were carried out via Google Cloud Platform. ADMET characteristics for borivilianoside H were determined using admetSAR web server. Results: Among the selected 7 top-ranked target proteins, fibroblast activation protein (FAP) exhibited the highest binding affinity followed by serum albumin (ALB), bone morphogenetic protein 2 (BMP2) and kinesin-like protein 11 (KIF11). However, the best fit was found with KIF11, where both the steroidal and oligosaccharide moieties of borivilianoside H were involved in interacting with the protein cavity. KIF11 was thus found to be the most suitable target for the anti-cancer effect of borivilianoside. ADMET analysis revealed its suitability as an intravenous drug. Conclusion: The targets predicted using this approach will serve as leads for the possible use of borivilianoside H, one of the active ingredients of Chlorophytum borivilianum as an anti-cancer drug.