Visualization of polyphenol distribution in persimmons using fluorescence fingerprint imaging

Abstract

Abstract Polyphenols are vital functional compounds, yet their spatial distribution remains difficult to quantify without destructive extraction. This study aimed to visualize polyphenol distribution in persimmon cross-sections using fluorescence fingerprint (FF) imaging, a technique that measures fluorescence intensities at multiple combinations of excitation and emission wavelengths. First, critical wavelengths for visualizing polyphenols were determined based on the loadings of a partial least squares regression model that estimated extractable polyphenol content (EPP) from FF data. The wavelength ranges Ex. 260–350 nm and Em. 330–440 nm were selected, and fluorescence images of five persimmon cultivars at immature and mature stages were acquired at these wavelength conditions. The loadings of principal component (PC) analysis matched the FF of epicatechin, with scores localized near the seeds and core. However, PC1 scores showed a negative correlation (r = − 0.29) with actual EPP content, appearing lower in the skin and immature fruits where EPP was higher. To understand this discrepancy, standard solutions of caffeic acid, quercetin, and epicatechin were mixed at different concentrations, and their FF and absorbance spectra were measured. The results showed that epicatechin fluorescence was re-absorbed by other co-existing polyphenols, such as caffeic acid and quercetin. The fluorescence intensity of the mixture is therefore suggested to reflect the interaction between different fluorescing and absorbing compounds rather than the concentration of total polyphenols. While FF imaging captured the spatial heterogeneity of fluorescent components, the results highlight the complexity of interpreting autofluorescence in intact tissues due to intermolecular interactions.