Alfa-mangostin is the major component in Mangosteen (Garcinia mangostana Linn) pericarp having several pharmacological activities including reducing blood pressure, antidiabetic, anticancer, and antioxidants. The objective of this study was to develop FTIR spectroscopy-multivariate calibration of partial least square (PLS) for quantitative analysis of alfa-mangostin and to classify mangosteen pericarp using principal component analysis (PCA). Mangosteen pericarps from different locations (Java provinces and South Sulawesi, Republic of Indonesia) were extracted using ethanol and were subjected to high performance liquid chromatography (HPLC) for analysis of alfa-mangostin and Fourier transform infrared (FTIR) spectroscopy measurements. HPLC was used to determine the levels of alfa-mangostin, used as actual values during FTIR spectroscopy analysis. The prediction of alfa-mangostin was obtained from the correlation between actual values and FTIR predicted values, facilitated with PLS model. The results showed that wavenumbers region of 3825-937 cm-1 offered reliable model with coefficient correlation (r) value of 0.9927 and root meat square error of calibration (RMSEC) of 0.0831%. The validation models also exhibited the accurate and precise results for prediction of alfa-mangostin with r-value of 0.9754 and root mean square error of prediction (RMSEP) value of 0.174%. Furthermore, the chemometrics of principal component analysis using variables of absorbances at selected fingerprint (1000-800 cm-1) could classify mangosteen pericarp from different regions. FTIR spectroscopy combined with chemometrics offered the reliable method for quality assurance of mangosteen pericarp.
Key words: alfa-mangostin, partial least square, principal component analysis, mangosteen pericarp, FTIR spectroscopy
|