An in silico approach to identify potential medicinal plants for treating Alzheimer disease: a case study with acetylcholinesterase.

Alzheimer's disease (AD) is a progressive neurological disorder affecting an estimated 10 million people worldwide. There is no cure for AD, and only a handful of drugs are known to provide some relief of the symptoms. The prescription drug donepezil has been widely used to treat to slow the progression and onset of the disease; however, the unpleasant side effects have paved the way to find alternative medicines. Many herbs are known to improve brain function, but evidence of medicinal plants that can treat AD is limited due to the lack of concrete rational evidences. Moreover, the traditional method of randomly screening plant extract against AD targets takes time and resources. In this study, a receptor-based in silico method has been implemented which serves to accelerate the process of identification of medicinal plants useful for treatment of AD. A database of natural compounds was compiled to identify hits against acetylcholinesterase (AChE). Receptor-based pharmacophore screening was performed, and selected hits were subjected to docking and molecular dynamics simulations. Molecular Mechanics/Generalized Born surface area (MM/GBSA) calculations were carried out to identify the best scoring hits further. In vitro assays were done for the plant extracts containing the top-scoring hits against AChE. Three plant extracts showed favorable inhibitory activity.Communicated by Ramaswamy H. Sarma.

MedPServer: A database for identification of therapeutic targets and novel leads pertaining to natural products.

Natural products have been the source of treatment for various human diseases from time immemorial. Interests in natural product-based scaffolds for the discovery of modern drugs have grown in recent years. However, research on exploring the traditional medicinal systems for modern therapeutics is severely limited due to our incomplete understanding of the therapeutic mechanism of action. One possible solution is to develop computational approaches, based on ligand- and structure-based screening tools, for fast and plausible target identification, leading to elucidation of the therapeutic mechanism. In the present work, we present two methods based on shape-based and pharmacophore search to predict targets of natural products and elucidate their mechanism, and to identify natural product-based leads. These methods were tested on an in-house developed database of medicinal plants that include information from a largely unexplored North-East region of India, known as one of the twelve mega biodiversity regions. However, depending on the choice of the lead molecules, any existing databases can be used for screening. MedPServer is an open access resource available at http://bif.uohyd.ac.in/medserver/.