Predicting cyclooxygenase inhibition by three-dimensional pharmacophoric profiling. Part II: Identification of enzyme inhibitors from Prasaplai, a Thai traditional medicine

Abstract

Prasaplai is a medicinal plant mixture that is used in Thailand to treat primary dysmenorrhea, which is characterized by painful uterine contractility caused by a significant increase of prostaglandin release. Cyclooxygenase (COX) represents a key enzyme in the formation of prostaglandins. Former studies revealed that extracts of Prasaplai inhibit COX-1 and COX-2. In this study, a comprehensive literature survey for known constituents of Prasaplai was performed. A multiconformational 3D database was created comprising 683 molecules. Virtual parallel screening using six validated pharmacophore models for COX inhibitors was performed resulting in a hit list of 166 compounds. 46 Prasaplai components with already determined COX activity were used for the external validation of this set of COX pharmacophore models. 57% of these components were classified correctly by the pharmacophore models. These findings confirm that the virtual approach provides a helpful tool (i) to unravel which molecular compounds might be responsible for the COX-inhibitory activity of Prasaplai and (ii) for the fast identification of novel COX inhibitors.

Fig. 1

Number of VH obtained from PS (grey columns) vs. number of known components of the plants Prasaplai is composed of (white columns).

Fig. 2

Decision tree for validation of pharmacophore model set.

Fig. 3

General workflow of the virtual PS approach performed in this study.

Fig. 4

Numbers of VH included in pharmacophore models validation. Threshold: highly active, IC₅₀ < 25.0 μM (dark grey); moderately active, IC₅₀ = 25.0–150.0 μM (light grey); inactive, IC₅₀ > 150.0 μM (white).

Fig. 5

Numbers of non-VH included in pharmacophore models validation. Threshold: highly active, IC₅₀ < 25.0 μM (dark grey); moderately active, IC₅₀ = 25.0–150.0 μM (light grey); inactive, IC₅₀ > 150.0 μM (white).