Abstract: In this study, the metabolite characteristics of six major pomegranate varieties grown in Huili were analyzed using ultra-high performance liquid chromatography-quadrupole time-of-flight-tandem mass spectrometry (UPLC-Q-TOF-MS/MS) combined with feature-based molecular networking (FBMN). Partial least square-discriminant analysis (PLS-DA) was used to screen for differential metabolites among cultivars, and a correlation matrix between these metabolites and fruit color and sensory characteristics was constructed to identify the chemical basis of cultivar differences. The results showed that a total of 111 metabolites were identified, including phenylpropanoids and polyketides (37.80%), lipids and lipid-like molecules (23.42%), organic acids and derivatives (12.61%), organic oxygen compounds (8.11%), benzenoids (7.21%), nucleosides and nucleotide derivatives (7.21%), and other compounds (3.60%). Furthermore, a total of 68 differential metabolites were selected, among which the ion intensities of cyanidin-3,5-di-O-glucoside and other pigments in ‘Jinzhuxiangfei’ and ‘Tunisian soft-seed’ were high and positively correlated with fruit redness. In addition, the ion intensities of sweetness-related organic oxygen compounds were high in ‘Tunisian soft-seed’ and ‘Jinzhuxiangfei’, while ‘Israeli sour’ was rich in citric acid and other sour and astringent components. The ion intensities of flavonoids such as luteolin-4’-O-glucoside and quercetin-4’-O-glucoside in ‘Jinzhuxiangfei’ were high, indicating potential bioactivity. These findings provide data support for the breeding and development of characteristic functional pomegranate varieties.

Key words: pomegranate; metabolomics; sensory properties; differential metabolites; ultra-high performance liquid chromatography-quadrupole-time-of-flight tandem mass spectrometry; feature-based molecular networking