Exploring different sample preparation techniques for NMR-based metabolomics analysis of Eugenia uniflora L. leaves.

Eugenia uniflora L. belongs to the Myrtaceae family and is popularly known as pitangueira. It is a native Brazilian tree, widely distributed throughout South America. Traditionally, its leaves have been utilized for treating ailments such as diarrhoea, stomach pains, worm infestations, hyperlipidaemia and hypertension. The variability in the composition of secondary metabolites in E. uniflora may be influenced by factors such as fruit color-based varieties and seasonal influence1. Changes in phenolic compound composition have also been demonstrated based on tree maturation. These variability changes can only be observed in metabolomic studies with high-quality data. Numerous studies have shown that sample preparation can significantly influence the metabolomic profile, potentially leading to misinterpretation of the data2. Considering the importance of analyzing the influence of chemical variability among different trees, this work aims to evaluate three different sample preparation techniques to preserve most of the metabolic information of the leaves from E. uniflora, as well as to identify the most discriminant compounds among the sample preparation methods using metabolomic-based NMR. The sample preparation procedures comprised three methods: (1) flash freezing in liquid nitrogen followed by freeze-drying (2) natural drying at room temperature and (3) dried at the oven for 12 hours at 60ºC. For NMR analysis, dried leaves were extracted with 1:1 CD3OD/KH2PO4 buffer in D2O with 0.1% of TSP (pH 6), and spectra were recorded on a Bruker Avance III (400 MHz for 1H) equipped with a BBI probe head using the noesyphpr1d pulse sequence, relaxation delay (d1= 2 s), acquisition time (AQ= 2.56 s), spectral width (SW= 16 ppm) and a number of scans (ns= 128). The NMR results showed that the sample preparation methods have a strongly influence on the sugar composition, mostly by promoting degradation of the disaccharides content (Figure 1). These preliminary results indicate that the sample method that most prevent data loss is instantly freezing the leaf in liquid nitrogen then freeze-dried the sample, after that treatment sample can be kept in - 80ºC until the moment of the analysis. These results drive the sequence of this work witch it may set the chemical difference of the varieties of E.uniflora trees.