Elisa Fierini

Characterisation of the Polyphenol Content in the Kiwifruit (Actinidia deliciosa) Exocarp for the Calibration of a Fruit-sorting Optical Sensor

INTRODUCTION Kiwifruit contains high amounts of anti-oxidants beneficial to health. Its quality is influenced by ripening time, genotype, cultivation techniques, climate and storage conditions after harvest. OBJECTIVE The aim of the present study was to characterise the phenolic content by HPLC methods and to evaluate the performance of a portable optical sensor (Multiplex 3), for in vivo non-destructive phenolic compound assessment in kiwifruits. METHODS Kiwifruits peel extracts were characterised by reverse-phase (RP) HPLC with diode-array detector (DAD) and electrospray ionisation (ESI) with MS using the Zorbax SB-Aq. column from Agilent. The fluorimetric sensor method is based on the screening of fruit chlorophyll fluorescence excitation and allows the UV absorbance of intact fruit skin to be measured. The flavonol index, FLAV, was calculated as log(FRF(R)/FRF(UV)), where FRF(R) and FRF(UV) are the chlorophyll fluorescence excited with red and UV light. RESULTS Hydroxycinnamic acids, procyanidins, and quercetin glycosides were the main polyphenol classes detected by HPLC-DAD-ESI/MS in the kiwifruit skin. A good linear regression (R²  = 0.88) was found between the fluorimetric sensor FLAV index and flavonol chromatographic analysis of the fruits. The FLAV index was able to detect the higher content of flavonols in sun-exposed fruits with respect to mid-shaded and shaded ones in accordance with the destructive analysis. CONCLUSION The fluorimetric sensor represents a rapid and non-invasive tool to: (i) monitor the flavonol accumulation in kiwifruit and to assess its quality concerning the healthy anti-oxidant properties; (ii) evaluate the effect of environmental and agronomical factors related to the fruit quality; and (iii) select fruits with the largest flavonol content, and consequently less susceptible to pathogen attack, in order to improve their storage durability.

Management Zone Delineation for Winegrape Selective Harvesting Based on Fluorescence-Sensor Mapping of Grape Skin Anthocyanins

We analyzed the potential of non-destructive optical sensing of grape skin anthocyanins for selective harvesting in precision viticulture. We measured anthocyanins by a hand-held fluorescence optical sensor on a 7 ha Sangiovese vineyard plot in central Italy. Optical indices obtained by the sensor were calibrated for the transformation in units of anthocyanins per berry mass, i.e., milligrams per gram of berry fresh weight. A full protocol for optimal data filtration, interpolation, and homogeneous zone delineation based on a very large number of optical measurements is proposed. Both the single signal-based fluorescence index (ANTHR) and the two signal ratio-based index (ANTHRG) can be used for Sangiovese grapes. Significant separations of grape-quality batches were obtained by several methods of data classification and zone delineations. Basic statistical criteria were as efficient as the K-means clustering. The best separations were obtained for three classes of grape skin anthocyanin.

Prediction models for assessing anthocyanins in grape berries by fluorescence sensors: Dependence on cultivar, site and growing season

Fluorescence sensors are useful tools for the non-destructive assessment of grape berry anthocyanins. The Multiplex (Mx) sensor here studied provides two anthocyanin indices: ANTHR = log(1/Chl-fluorescence_R) and ANTHRG = log(Chl-fluorescence_R/Chl-fluorescence_G), based on the chlorophyll (Chl) fluorescence excited with red (R) and green (G) light. These indices were calibrated against wet chemistry. The dependence of anthocyanin prediction models on cultivar, season and site was studied on four cultivars in two Italian regions during three consecutive years. The 2010 global model (all cultivars at both growing sites) gave relative prediction errors on anthocyanin content less than 14.1% (ANTHR) and 19.0% (ANTHRG). The ANTHRG was independent of season, maintaining a relative error of about 20% in both 2011 and 2012. In field applications of the calibrated Mx, it showed its ability to detect inter-plot and inter-season differences on both growing sites.