Environmental Science
E. Imelda; K. Khairan; R.R. Lubis; T. Karma; R. Idroes
Abstract
BACKGROUND AND OBJECTIVES: Isotoma longiflora L is commonly used as a medicinal plant by the local community in Indonesia, and its geographical position determines its bioactive compounds and hence its efficacy. Ethanol extracts of Isotoma longiflora leaves from various locations in Aceh Province were ...
Read More
BACKGROUND AND OBJECTIVES: Isotoma longiflora L is commonly used as a medicinal plant by the local community in Indonesia, and its geographical position determines its bioactive compounds and hence its efficacy. Ethanol extracts of Isotoma longiflora leaves from various locations in Aceh Province were analyzed using a simple infrared spectroscopy technique combined with chemometrics to determine the effect of geographical location and conditions by classification and authentication.METHODS: Isotoma longiflora leaf samples were collected from Aceh Besar (a geothermal manifestation of Ie Suum), Banda Aceh, Aceh Jaya, Bireun, and Central Aceh. Principal component analysis was used to categorize the ethanol extract of Isotoma longiflora leaves, and a linear discriminant analysis was used for authentication.FINDINGS: The principal component analysis score plot indicated 89 percent of total data variance and that the samples formed three distinct groups: group I consisting of Aceh Tengah and Bener Meriah samples; group II of Aceh Besar and Banda Aceh samples; and group III of Aceh Selatan, Aceh Barat Daya, Aceh Jaya, and Bireun. A linear discriminant analysis was then used to validate these results, and the linear discriminant analysis model derived from the cross-validation predicted the origin of Isotoma longiflora samples with 100 percent accuracy rate.CONCLUSION: The Isotoma longiflora leaf extracts were successfully classified using Fourier-transform infrared spectroscopy data processed through chemometric calculations (namely, principal component analysis). Based on the cross-validation using linear discriminant analysis showed that the prediction model had a 100 percent accuracy. The present study thus revealed the effect of geographical location on the composition of bioactive compounds in Isotoma longiflora, suggesting the potential of chemometric techniques for quality control and assurance in traditional medicine.
