Metabolic Profiling and Predicting the Free Radical Scavenging Activity of Guava (Psidium guajavaL.) Leaves According to Harvest Time by¹H-Nuclear Magnetic Resonance Spectroscopy

Abstract

Guava leaves were classified and the free radical scavenging activity (FRSA) evaluated according to different harvest times by using the (1)H-NMR-based metabolomic technique. A principal component analysis (PCA) of (1)H-NMR data from the guava leaves provided clear clusters according to the harvesting time. A partial least squares (PLS) analysis indicated a correlation between the metabolic profile and FRSA. FRSA levels of the guava leaves harvested during May and August were high, and those leaves contained higher amounts of 3-hydroxybutyric acid, acetic acid, glutamic acid, asparagine, citric acid, malonic acid, trans-aconitic acid, ascorbic acid, maleic acid, cis-aconitic acid, epicatechin, protocatechuic acid, and xanthine than the leaves harvested during October and December. Epicatechin and protocatechuic acid among those compounds seem to have enhanced FRSA of the guava leaf samples harvested in May and August. A PLS regression model was established to predict guava leaf FRSA at different harvesting times by using a (1)H-NMR data set. The predictability of the PLS model was then tested by internal and external validation. The results of this study indicate that (1)H-NMR-based metabolomic data could usefully characterize guava leaves according to their time of harvesting.

The potential of metabolic fingerprinting as a tool for the modernisation of TCM preparations

“…1 H NMR analysis has been used for quality control (Tarachiwin et al, 2008;Lee et al, 2009;Frédérich et al, 2011), identification of compounds, impurities and adulterants (Peng et al, 2004;Balayssac et al, 2009a,b;Holzgrabe and Malet-Martino, 2011), quantification (Pauli et al, 2005), determination of species origin (Kang et al, 2008) and correlation of a plant's metabolome with biological activity (Tarrago et al, 2006;Kim et al, 2011). Hyphenated techniques such as LC-NMR have found application in the analysis of plant extracts, wherein metabolites are separated and enter the NMR instrument directly from the HPLC column so that no isolation and sample manipulation is required (Wolfender et al, 2001).…”

“…Frutos et al (2006) have also used 19 F NMR for the evaluation of the inhibitory activity of new compounds against HIV-1 protease. More recently, analysis of 1 H NMR data sets by PCA and partial least square discriminant analysis (PLS-DA) have shown clear correlations between the free radical scavenging activity of Psidium guajava L. (Kim et al, 2011) and the anxiolytic and sedative activity of Galphimia glauca Cav. with their respective metabolomes (Cardoso-Taketa et al, 2008).…”

“…the results of a bioactivity assay. Examples for this procedure have been published for anxiolytic and sedative activity of Galphimia glauca (Cardoso-Taketa et al, 2008) and antioxidant activity of Psidium guajaval (Kim et al, 2011).…”

Comparison of partial least squares and random forests for evaluating relationship between phenolics and bioactivities of Neptunia oleracea

“…One of the regression models that is commonly used in these approaches is partial least squares (PLS). The PLS model has been successfully applied to identify the potential active metabolite corresponding to several biological activities, including antioxidant, antiproliferative, anti‐inflammation, α‐glucosidase inhibition and adenosine A1 receptor binding activities. Recently, a machine learning method, random forests, has drawn increased popularity as an application of MVDA tools in omics research.…”

Chemical characterization and antioxidant capacity of the araçá-pera ( Psidium acutangulum ): An exotic Amazon fruit

“…5) (7) was isolated from the hexane extract of the fruits of Psidium cattleianum (McCook-Russell et al, 2012). Citric and malic acids have been identified in fruits of P. guajava by 1 H NMR (Kim et al, 2011) and HPLC-UV (Wilson et al, 1982).…”