Taste of medicinal plants: A potential tool in predicting ethnopharmacological activities?

Ethnopharmacological relevance: Taste (rasa) is traditionally valued in Ayurveda as an important ethnopharmacological category, and reported to correlate with certain therapeutic activities (karman). The present paper endeavors to provide for the first time a type of mapping of ayurvedic ethnopharmacological space by the integration of medicinal plant taste-activity relationship (TA) data, originating from several traditional sources. Material and methods: The present TA database included 183 ayurvedic medicinal plants, with their traditional description of taste and ethnophamacological actions (EPA). 111 plants had a unique taste, and 72 plants a combination of several tastes, consisting of one principal taste, and other 2-4 secondary tastes. 121 ethnopharmacological actions were considered. Potential global or individual associations between rasa and karman were statistically analyzed. Results: We found a statistically significant global correspondence between the predominant plant taste and the ethnopharmacological activities (Chi-square test, p < 0.05), and the following individual associations (Fisher exact test, p < 0.05, positive association if ln OR > 0, negative association if ln OR < 0): bitter- 11 positive associations (ln OR 0.940 to 3.544) and 8 negative associations (ln OR -0.997 to -3.101), sweet- 14 positive associations (ln OR 0.839 to 3.132) and 20 negative associations (ln OR -0.962 to -3.270), pungent- 23 positive associations (ln OR 0.835 to 4.126) and 11 negative associations (ln OR -1.255 to -3.147), astringent- 7 positive associations (ln OR 1.099 to 3.813) and 7 negative associations (ln OR -1.076 to -2.631), sour- 7 positive associations (ln OR 1.128 to 4.664) and 1 negative association (ln OR -2.244). Among the 109 statistically significant associations, 24 were not traditionally mentioned, and only 4 were in opposition with the traditional ayurvedic sources. Discussions: The present study confirms the traditional relationship rasa-karman and demonstrates that statistical models can be used to predict EPAs of medicinal plants depending on their taste. Integration of the ayurvedic concept of taste into ethnopharmacological research might reveal new facets of plant therapeutic potentials and provide a framework for developing a probabilistic approach to drug discovery that can be exploited to increase bioprospecting research efficacy. (C) 2015 Elsevier Ireland Ltd. All rights reserved.