Intergrated Sensory Evaluation With Metabolomics To Compare The Quality Differences Of Citrus Fruits From Different Regions

Fruit sensory quality is the most important factor to determine the competitiveness of the fruit market.In order to scientifically evaluate the differences of citrus fruit sensory quality,nine ‘Ehime Kashi No.28’ samples,nine ‘Orah’ mandarin samples and four Sour Free Sweet Orange samples from different regions were used as research materials in this study.Firstly,the evaluators were organized to evaluate each sensory attribute and weight of the samples,and the fuzzy mathematics was used to derive the ranking of the weights of different sensory attributes and to rank and group the comprehensive sensory qualities of samples of different regions;meanwhile,conventional physical and chemical indexes were detected,and metabolites related to fruit flavor and color were analyzed by GC-MS and HPLC,including soluble sugars,organic acids,amino acids,volatile substances,flavonoids and carotenoids;following that,OPLS-DA was used to analyze the main differential metabolites of the samples between different groups;finally,each sensory attribute of the samples was correlated with the relevant metabolite data by Pearson correlation analysis.The main results are shown as follows:1.Sensory quality differences of ‘Ehime Kashi No.28’ from different regionsThe results of sensory evaluation and fuzzy ranking showed that the ranking of sensory attributes weight of ‘Ehime Kashi No.28’ was pulp sweetness > pulp acidity >pulp mastication > fruit aroma > fruit color = fruit shape.The high sensory quality group were samples from regions 9,3,2,8 and 7,and the low sensory quality group were samples from regions 4,5,1 and 6.OPLS-DA analysis showed that six metabolites significantly differentiated taste differences between the high sensory quality and low sensory quality groups,including malic acid,L-aspartic acid,α-ketoglutaric acid,D-(+)-glucose,L-proline,and D-fructose.Pearson correlation analysis showed thatβ-cryptoxanthin and violaxanthin contents were significantly positively correlated with fruit color,D-fructose,D-(+)-glucose and D-(+)-tarose contents were significantly positively correlated with pulp sweetness,and malic acid and L-asparagine contents were significantly negatively correlated with pulp sweetness,the above substances were the signature metabolites reflecting the comprehensive sensory quality of ‘Ehime Kashi No.28’.2.Sensory quality differences of ‘Orah’ Mandarin from different regionsThe ranking of sensory attributes weight of ‘Orah’ Mandarin was pulp sweetness >pulp acidity > pulp mastication > fruit color > fruit aroma > fruit shape.The high sensory quality group included samples from regions 1,4,6,9,3 and 5,and the low sensory quality group included samples from regions 2,7 and 8.OPLS-DA analysis showed that nine major differential metabolites significantly differentiated the high and low sensory quality groups in terms of taste,including L-glutamic acid,hesperidin,L-valine,citric acid,α-ketoglutaric acid,γ-aminobutyric acid,allo-inositol,sucrose,and malic acid.Pearson correlation analysis showed that α-ketoglutaric acid,L-glutamic acid and hesperidin content were significantly negatively correlated with pulp sweetness,which were the signature metabolites reflecting the comprehensive sensory quality of ‘Orah’Mandarin.3.Sensory quality differences of Sour Free Sweet Orange between Yunnan and Hunan regionsSensory evaluation results showed that Xinping-1 and Xinping-2 samples had better mastication,aroma and overall quality compared to Hunan ‘Bingtang’ Navel Orange and Hunan ‘Bingtang’ Sweet Orange,and Xinping-1 had better taste but lighter peel color.Metabolomic and physicochemical analyses showed the sucrose and D-(+)-xylose contents of Xinping-1 pulp was significantly higher than that of the two samples from Hunan(P < 0.05,the same as below),while the citric acid content was the opposite;the pulp hardness of Xinping-1 and Xinping-2 was significantly lower than that of the two samples from Hunan;the types of volatile substances in the peels of the four samples were the same,but nearly 80% of the volatiles,the content of which was significantly higher in Xinping-1 and Xinping-2.Based on the criteria of P < 0.05 and VIP > 1,36 differential metabolites were screened,among which only L-proline and(E)-2-hexenal were high in two samples from Hunan,and all other components were dominant in samples from Yunnan,including γ-aminobutyric acid,β-copaene,α-copaene,and valencene,etc.This study combined fuzzy modeling of sensory scores and metabolomics,aiming to establish a comprehensive sensory quality evaluation method for citrus fruits,clarify the differences in sensory quality of citrus fruits from different regions,explore consumer preferences,reveal metabolites closely related to citrus sensory attributes,and provide theoretical support and digital methods for citrus breeding,practical production and marketing to improve citrus quality.