| Summary: | Rocha pear (Pyrus communis L. var. Rocha) is a traditional Portuguese variety of pear recognized by its organoleptic qualities. Adding to the known benefits of consuming fruit in general, pears are considered a good source of fiber and potassium with low sodium content, combined with low glycemic index (Barda, 2011; Foster-Powell et al., 2002) . The nutritional value of pears, in particular its mineral content, is of great relevance for human nutrition and public health. In this sense, Food Composition Databases (FCDBs) are a crucial tool. In this study, we analyze the mineral content of Rocha pears to promote the linkage between the nutritional profile and geographical origin, filling the gap of absent values in FCDBs. Rocha pears produced in the west part of Portugal can be certified as Protected Designation of Origin (PDO), under the label of “Pera Rocha do Oeste”. These represent the largest PDO production in the country (Globalagrimar, 2016). However, this variety is also cultivated in other areas at national level (nPDO) and is gaining interest at international level due to growing exportations. The mineral profile of plants is a reflection of the soil type and the environmental conditions under which plants were grown. Therefore, minerals and trace elements, as well as their isotopes, are expected to be suitable biomarkers in authenticity studies (Coelho et al., 2017; Gonzalvez et al., 2009). Nonetheless, to extract the most information from datasets these must to be combined with chemometric tools capable of performing multivariate data analysis. The present work aims to develop a chemical fingerprint of Rocha pears produced within the PDO region based on multielement analysis and isotope ratios (IR). For such purpose 150 pear samples were collected from 10 farms located in two regions (PDO and nPDO). Samples were analyzed in pools, each composed of 3 pears from the same tree, leading to 50 pools. Multielement analyses were carried out, in triplicate, in compliance with NP EN ISO/IEC 17025 and EN 1380. The content of twenty-four minerals and trace elements was measured by 2 inductively coupled plasma optical emission spectrometer (ICP-OES) (Fe, Mg, P, Ca, Na, K) and ICP mass spectrometer (ICP-MS) (Li, Be, B, Al, Cr, Mn, Co, Ni, Cu, Zn, As, Se, Sr, Mo, Cd, Sn, Tl, Pb). A Multicollector-ICP-MS (MC-ICP-MS) was used to determine Sr IR. Results showed a low level of the analyzed contaminants indicating that these do not present a risk for consumers. K was the most abundant element, followed by P. Be, Se, Sn and Pb were systematically below the limit of quantification (LOQ) and were not included in the statistical treatment. The dataset for multielement content and Sr IR was analyzed resorting to chemometric techniques, namely Principal Component Analysis (PCA), Hierarchical Clustering Analysis (HCA) and Linear Discriminant Analysis (LDA). Differences were identified in the level of inorganic components between the regions under study. The most significant were in the levels of Al, Ni and Na, which were higher in samples originating from the PDO region, while the content of Sr and Tl were superior outside this region. The exploratory analysis, based on the unsupervised models PCA and HCA, identified relationships between samples and their geographical provenance. LDA confirmed the possibility of tracing pear samples to the correspondent origin based on mineral profile and Sr IR. Samples used for cross validation of the predictive model were matched to its origin with 100% accuracy. In conclusion, the analytical methods were suitable for the purpose and are recommended methods for the analysis of inorganic components even at low values. The chemometric techniques used allowed the differentiation of samples according to their geographical origin, contributing to its authenticity and traceability as well as to a deeper knowledge of the mineral content of Rocha pear. The combination of these two techniques is a proper tool for mineral fingerprinting of Rocha Pear’s geographical origin.
|
|---|